International journal of computer integrated manufacturing , tập 24, số 1, 2011

Keywords: CAPP; machining process; features; planning; operation 1.. In the domain of machining processes, which is the focus of this article, the major process planning activities may i

Computer-aided process planning – A critical review of recent developments and future trends

Xun Xua*, Lihui Wangb and Stephen T Newmanca

Department of Mechanical Engineering, School of Engineering, University of Auckland, Private Bag 92019, Auckland 1142,New Zealand;bVirtual Systems Research Centre, University of Sko¨vde, P.O Box 408, 541 28 Sko¨vde, Sweden;cDepartment of

Mechanical Engineering, University of Bath, Bath, BA2 7AY, United Kingdom(Received 28 January 2010; final version received 22 August 2010)For the past three decades, computer-aided process planning (CAPP) has attracted a large amount of researchinterest A huge volume of literature has been published on this subject Today, CAPP research faces new challengesowing to the dynamic markets and business globalisation Thus, there is an urgent need to ascertain the currentstatus and identify future trends of CAPP Covering articles published on the subjects of CAPP in the past 10 years

or so, this article aims to provide an up-to-date review of the CAPP research works, a critical analysis of journalsthat publish CAPP research works, and an understanding of the future direction in the field First, generalinformation is provided on CAPP The past reviews are summarised Discussions about the recent CAPP researchare presented in a number of categories, i.e feature-based technologies, knowledge-based systems, artificial neuralnetworks, genetic algorithms, fuzzy set theory and fuzzy logic, Petri nets, agent-based technology, Internet-basedtechnology, STEP-compliant CAPP and other emerging technologies Research on some specific aspects of CAPP isalso provided Discussions and analysis of the methods are then presented based on the data gathered from theElsevier’s Scopus abstract and citation database The concepts of ‘Subject Strength’ of a journal and ‘technologyimpact factor’ are introduced and used for discussions based on the publication data The former is used to gaugethe level of focus of a journal on a particular research subject/domain, whereas the latter is used to assess the level ofimpact of a particular technology, in terms of citation counts Finally, a discussion on the future development ispresented

Keywords: CAPP; machining process; features; planning; operation

1 Introduction

Design and manufacturing is a critical phase of

product development process The pivotal link

be-tween design and manufacturing is process planning

Process planning deals with the selection of necessary

manufacturing processes and determination of their

sequences to ‘transform’ a designer’s ideas (namely the

designed part) into a physical component economically

and competitively In the domain of machining

processes, which is the focus of this article, the major

process planning activities may include interpretation

of design data, selection of machining operations,

machine tools, cutting tools, datum and fixture as well

as calculation of cost and production time No doubt,

this is a complex engineering problem

The traditional approach to solving

process-plan-ning problems in a manufacturing company is to leave

it in the hands of manufacturing experts These

domain experts use their experience and knowledge

to generate instructions for the manufacture of

products based on design specifications and available

facilities Different process planners often come up

with different plans for the same problem, addinginconsistency to the already complicated problem.Since Niebel (1965) first discussed the use ofcomputers to assist process-planning tasks, morethan 40 years have elapsed In comparison withcomputer-aided design (CAD) and computer-aidedmanufacturing (CAM), computer-aided process plan-ning (CAPP) has been lagging behind in terms ofproviding practical, matured, professional and com-mercialised solutions to the manufacturing industry.This is though not attributed to the lack of researcheffort On the contrary, there is a prolonged andprolific history of research and publications

In general, there are two approaches in CAPP,variant and generative Variant approach follows theprinciple that similar parts require similar plans Itrequires a human operator to classify a part, input partinformation, retrieve a similar process plan from adatabase and make the necessary modifications Thisapproach is suited to enterprises involving stablemanufacturing processes and manufactured productsthat vary little The advantage of this approach is theease of maintenance, but the shortcoming is the lack of

*Corresponding author Email: x.xu@auckland.ac.nz

Vol 24, No 1, January 2011, 1–31

ISSN 0951-192X print/ISSN 1362-3052 online

Ó 2011 Taylor & Francis

DOI: 10.1080/0951192X.2010.518632

an on-time calculation of manufacturing process, and

quality of the process plan still depends on the

knowledge of a process planner Manual input is still

required to establish the mass data of manufacturing

processes In a generative approach, process plans are

generated with little human intervention It generates

new process plans by means of decision logic and

process knowledge Earlier developments of this

approach adopted decision-making to determine the

manufacturing process and use group technology code

or special descriptive languages to define workpieces

Later development (since the mid 1980s) focused on

the use of features to define product models The

bottleneck of this approach is the difficulty in

obtaining useable features, and the difficulty in

representing, managing and utilising human expertise

This is the main reason that both feature technology

and knowledge-based techniques have been heavily

researched in association with CAPP

Some of the desirable characteristics of an effective

CAPP system are to

Be interconnected with up- and down-stream

activities, i.e design and manufacturing, in such

a way that a CAPP system can take design data

as it is and generate output that can be fed into a

CAM and later a CNC system;

Be extendible, adaptable and customisable for

individual enterprises and to new processes;

Provide effective knowledge acquisition,

repre-sentation and manipulation mechanisms as well

as the means to check the completeness and

consistency of that knowledge;

Involve users in some parts of the

decision-making process, provide heuristics as needed and

supplement the system’s abilities; and

Come with a user-friendly interface in support of

effective interaction by facilitating inputs,

produ-cing outputs and reports, and displaying the

results graphically

Because the concept of CAPP was suggested, there

have been numerous research publications as well as

technical surveys It is also evident that the trend of

CAPP research has also undergone drastic changes To

the authors’ knowledge, there has not been a

compre-hensive review of the CAPP research for machining

since 1998 Therefore, the aim of this article is to

provide a comprehensive review on CAPP technologies

developed for machining since the late 1990s but

mostly after 2000 It is also to be pointed out the

ontology and architecture matters are not included in

this review There are seven sections in the article

Section 2 provides an excerpt for the past review

publications on CAPP Section 3 is the main section

describing various technologies developed or mented These include the feature-based technologies,knowledge-based systems, neural network, geneticalgorithm (GA), fuzzy set theory/logic, Petri net(PN), agent-based technologies, Internet-enabledCAPP, STEP-compliant CAPP as well as someemerging technologies Section 4 comments on theresearch work on some specific areas of CAPP, e.g.tool section, setup planning, operation selection andsequencing, decision models and integrated processplanning and knowledge representation Discussionsand future trends in CAPP are presented in Sections 5and 6, respectively Section 7 concludes the article

imple-2 Previous reviews in CAPPThe idea of developing process plans using computerswas first presented in 1965 by Neibel (1965), and thefirst CAPP system was developed in 1976 under thesponsorship of Computer Aided Manufacturing Inter-national (CAM-I) (Cay 1997) Since then, there hasbeen a plethora of research work in the area of CAPP;

so too are a significant number of surveys This sectionprovides a snap-shot of the past surveys in the field.One of the first review articles was written bySteudel in 1984 The author discussed the approachesand strategies for structuring manufacturing methodsand data for the development of a generative type,automated planning system An overview of theinformation needed for such a task was provided.This article also outlined the anticipated development

of a ‘common language of geometry’ to relate a part tothe process, and development of CAD/CAM systemsthat incorporated CAPP In the following year, Ever-sheim and Schulz (1985) presented a survey based onquestionnaires sent to the CAPP developers and end-users in Europe, North America and several Asiancountries during 1983 and 1984 From the survey, itwas apparent that the CAPP development andapplications were still relatively new In 1988, Hamand Lu (1988) presented an evaluation of the status ofCAPP at the time, and correctly stated that thedirection of future research lies in the integrationbetween design and manufacturing, and the use ofartificial intelligence (AI) technologies Following onfrom that is probably the most significant survey of thetime, by Alting and Zhang in 1989 In this survey, theauthors reviewed over 200 published works andfeatured 14 well-known CAPP systems One hundredand fifty-six existing systems were listed in a tableformat The survey indicated the difficulty in integrat-ing CAD with CAPP due to a lack of commonmethods to represent geometric entities The authorsalso suggested the interfacing issues between CAPPand CAM and other computerised production systems

such as NC tool path, MRP, production simulation,

etc They recognised AI technologies as a crucial

technology in the development of an effective process

planning system In addition, the importance of

learning systems was pointed out, and an ideal

approach identified to integrate all the information

involved in production of a part into a single database

In the same year, Gouda and Taraman (1989)

published a survey of the 128 CAPP systems at the

time Four types of CAPP systems were highlighted,

variant, semi-generative, generative and expert

pro-cess-planning systems

The next survey was compiled by the CAPP

Working Group of the CIRP in 1993 (ElMaraghy

et al 1993) Aspects covered by this article include the

major development thrust in CAPP, the industry

perspectives of CAPP, evolving trends and challenges,

and integration of design, CAPP and production

planning Issues of quality and evolving standards

were also addressed In the same year, Eversheim and

Schneewind (1993) provided a broad but concise

review of CAPP in real industrial environments It

suggested that the future of CAPP development was an

extension to assembly planning, function integration

with NC programming, use of AI methods in

decision-making, and use of database sharing for data

integra-tion with CAD In 1995, Kamrani et al presented an

overview of the techniques and the role of process

planning It also discussed the critical issues and the

characteristics associated with evaluation and selection

of a CAPP system These issues are the range of

product support, classification / coding / graphic

capabilities, work instruction creation, process

plan-ning approaches, time analysis capabilities, machiplan-ning

parameters, material and tooling databases, system

requirements, cost, commercial availability and user

friendliness, vendor qualification and support

The next comprehensive review was written by

Leung (1996), where he compiled and annotated about

200 publications in CAPP from 1989 to 1996 The

author observed that solid modelling in CAPP systems

was not as adequate as anticipated, hence the

revitalisation of variant process planning systems

Leung believed that it was logical that future process

planning systems be built on intelligent system

architecture with AI techniques Following on from

Leung was the review by Cay and Chassapis (1997),

which covers the research work on CAPP from 1990 to

1997 It gave an overview of manufacturing features

and feature recognition techniques with CAPP

re-search Cay maintained that a fully automated

environment was not far from a reality, should

effective integration of design and manufacturing

systems be achieved The last general review on

CAPP in the 1990s is by Marri et al (1998) This

review covered the literature from 1989 to 1996 Theadvantages and disadvantages of these systems werediscussed with the generative approach highlighted.Aside from general CAPP reviews, there are alsosurveys in a more specific area, such as CAD andfeature-based process planning (Shah 1991, Shah et al.1991), neural network-based process planning (Yue et al.2002), expert system-based process planning (Gupta andGhosh 1989, Kiritsis 1995, Metaxiotis et al 2002a; Liao2005), and virtual reality-based process planning (Peng

et al 2000) More recently, Shen et al (2006a,b)presented a state-of-the-art survey on agent-based,distributed manufacturing process planning and sche-duling They addressed the complexity of manufacturingprocess-planning and scheduling problems, and reviewedthe literature in process planning, scheduling, and theirintegration, with a focus on agent-based approaches.Zhang and Xie (2007) provided a review on agenttechnology for collaborative process planning Keyissues in developing an agent-based process planningsystem were explored, including agent and systemarchitecture, communication standards and protocols,and applications

3 Current status of CAPPCentred on the CAPP technologies and systems, thissection consists of 10 sub-sections, each representing acategory to which the related technologies belong Thecategories are feature-based technologies, knowledge-based systems, artificial neural networks, GAs, fuzzyset theory and fuzzy logic, PNs, agent-based technol-ogy, Internet-based technology, STEP-compliantCAPP and other emerging technologies

3.1 Feature-based technologiesFeature technology has been the central topic forCAD/CAM integrations for years, so has it been forCAPP This is because almost all CAPP systemsfunction on the basis of features, or require features

to be the input data There are two approaches toobtaining features: feature recognition and design byfeatures (Shah 1991, Shah et al 1991) The featurerecognition approach examines the topology andgeometry of a part and determines the existence anddefinitions of features To achieve this, a geometricmodel of lower-level entities (lines, points, etc.) isconverted into a feature-based model in terms ofhigher-level entities (holes, pockets, etc.) Featurerecognition has been adapted in various approaches,such as rule-based approach, volume decompositionapproach, expert system, and graph-based approach.The design-by-feature approach builds a part frompredefined features stored in a feature library

Geometry of these features is defined but their

dimensions are left as variables to be instantiated

when a feature is used in the modelling process There

are two distinct methodologies for the

design-by-feature approach The first is destruction by machining

features and the other is synthesis by design features

Destruction by machining features method starts with

the model of the raw stock from which a part is to be

machined The design model is then generated by

subtracting depression features corresponding to the

material to be removed by machining operations from

the stock Synthesis by the design features method is

built by both adding and subtracting features

In a recent survey by Babic et al (2008), three

major feature recognition problems were identified, (i)

extraction of geometric primitives from a CAD model;

(ii) defining a suitable part representation for form

feature identification; and (iii) feature pattern

match-ing/recognition The review has a focus on the

rule-based methods It is perhaps fair to say that with some

exceptions, much of the focus of the feature

recogni-tion research has been on finding all possible features,

leaving the task of manufacturability analysis to

process planners Xu and Hinduja (1997, 1998)

recognised this issue and have developed methods for

recognising features specifically for machining

opera-tions such as roughing, semi-finishing and finishing

operations Han and Han (1999) proposed to integrate

feature recognition with process planning They used

feature recognition for manufacturing and setup

minimisation, feature dependence construction, and

generation of an optimal feature-based machining

sequence The system interacts with the tool database

and does manufacturability analysis together with

feature dependency construction In the same year,

Khoshevis et al (1999) published an integrated

process-planning system, using feature reasoning and

space search-based optimisation The process-planning

system consists of feature completion module, process

selection module and process sequencing module They

used graph-based methods and volume-based

meth-ods The features are recognised by an

object-orientated feature finder The feature completion

module creates a feature precedence network It also

allows the process planning system to accept the input

from some other feature recognizers Lee et al (2007)

developed a projective feature recognition algorithm

that outputs features that can be directly used for

process planning Process planning is based on the

topological sorting and breadth-first search of graphs

A great deal of the research work in the area of

machining feature recognition is limited to 2½ and 3

axis milling features (Sridharan and Shah 2004,

Ranjan et al 2005) In the work of Huang and

Yip-Hoi (2002), they suggested a methodology to extract

user-specific features from generic features This isachieved by specifying patterns for these specificfeatures High-level features are recognised and aremore meaningful to process planning Sadaiah et al.(2002) developed a generative CAPP system forprismatic parts The proposed system is divided intothree modules The first module is concerned withfeature extraction The second and third modules dealwith planning the set-up, machine selection, cuttingtool selection, cutting parameter selection, and gen-eration of process plan sheet Woo et al (2005)developed a hybrid feature recognizer for machiningprocess planning It is the integration of three distinctfeature recognition methods, i.e graph matching, cell-based maximal volume decomposition, and negativefeature decomposition Hou and Faddis (2006) in-vestigated the integration of CAD/CAPP/CAM based

on machining features In the system, machiningfeatures are utilised to carry machining geometryinformation from CAPP to CAM systems for thepreparation of tool paths Hwang and Miller (1997)described a process-planning model using mixed-typereasoning designed for processing prismatic parts onCNC machine tools in a batch-manufacturing envir-onment The mixed-type reasoning handles featureinteractions by combining forward chaining for featuresequencing and backward chaining for the construc-tion of a process plan, allowing the human problem-solving strategies to be decoupled from the tools foranalysis and sorting algorithms

Clearly, feature-based approaches have been widelyadopted by both CAD and CAPP systems (Patil andPande 2002) In the research by Markus et al (1997), afeature-based process planning system was developed,where planning of the framework is accomplished viaretrieving and adapting previous part family relatedplans The developed method can generate feasiblematching of different parts Another crucial issue inprocess planning is the sequence of machining processes.Using STEP AP224 features (ISO 10303–22 1999),Gonzalez and Rosado (2004) defined an internal featuremodel for process planning This way, all the informa-tion is represented around the machining feature forprocess planning without the use of geometric entities

In the past, operation sequencing received moreattention and has been studied deeply in variousaspects Wang et al (2006) presents a differentapproach as part of their distributed process planning(DPP) system As a two-layer hierarchy is considered

to separate the generic data from those that aremachine-specific in DPP, machining process sequen-cing is treated as machining feature sequencing withinthe context The advantage of their approach is thatboth manufacturing interactions and geometric inter-actions are handled during feature sequencing

One of the major hurdles in the development of a

comprehensive CAPP system is that each CAPP

domain is quite unique in terms of the analytical

models and knowledge bases it utilises As a result, the

field of CAPP has become highly fragmented (Yuen

et al 2003) A unifying theme seems to be needed to

reverse this trend One way of achieving this goal is to

adopt the paradigm of feature-orientation as the

unifying theme Yuen et al (2003) created a generic

CAPP support system (GCAPPSS) that can act as the

‘front end’ for all domain-specific CAPP systems The

GCAPPSS invokes a set of algorithms that enable

feature extraction, recognition, coding, classification

and decomposition The output from this system

enables a multi-layered hierarchical part representation

that can facilitate interpretation of feature relations

and the object itself

3.2 Knowledge-based systems

The art of process planning relies heavily on the

knowledge of experienced workers or domain experts

Hence, it is a knowledge intensive exercise An expert

system (also known as knowledge-based system) as

Welband (1983) stated in his work, ‘is a program which

has a wide base of knowledge in a restricted domain, and

uses complex inferential reasoning to perform tasks which

a human expert could do’ Undoubtedly,

knowledge-based systems have found numerous applications in

process planning An expert system usually consists of

three main components: the knowledge base, the

inference engine and the user interface Expert systems

have advantages over traditional computer systems,

since they organise knowledge in rules and control

strategies which allow users to modify a program with

ease, and they are able to organise knowledge in such a

way that they can reason intelligently Thus, expert

systems are able to deal with more complicated problems

such as process planning Also, export systems can be

designed so that they accumulate knowledge as time

passes, in the form of separate facts, production rules,

objects, etc The inference mechanism of an expert

system makes it possible to perform operations on the

knowledge base of analysed elements (Grabowik and

Knosala 2003) Park (2003) discussed the knowledge

capturing methodology in process planning To identify

the knowledge elements, three sub-models were

sug-gested: object model, functional model and dynamic

model, based on which three knowledge elements for

process planning were derived: facts (from the object

model), constraints (from the functional model), and

way of thinking and rules (from the dynamic model)

In process planning, an organised relationship

between design and manufacturing knowledge is

important The knowledge should be structured in

such a way that it allows for easier reasoning in thegeneration of a sound process plan In the research bySormaz and Khoshnevis (1995), a knowledge repre-sentation scheme that recognised both geometric andfeature-based representation of parts was proposed.The developed system can connect feature and processknowledge with part geometry, and use an object-oriented approach for a detailed presentation ofmachining knowledge Jia et al (2003) also adoptedthe object-oriented technology to represent setupprocess information, process decision knowledge anddecision procedure control knowledge In the sameyear, Grabowik and Knosala (2003) presented amethod of knowledge representation in the form of aset of objects Based on the set, the hierarchicalstructure of classes was prepared A more recentpublication by Denkena et al (2007) described aholistic process-planning model based on an integratedapproach combining technological and business con-siderations Halevi and Wang (2007) argued thatinstead of making decisions engineers should beengaged in the development of knowledge-based

‘road map’ The road map method can introduceflexibility and dynamics in the manufacturing processand thus simplifies the decision-making process inproduction planning Liu and Wang (2007) used ahybrid approach whereby knowledge-based rules andgeometric reasoning rules are combined to sort out thesequence of interacting prismatic machining features.Knowledge-based CAPP systems remain to be apopular branch of CAPP research since the 1980s.Anwer and Chep (1999) created an Intelligent ProcessPlanning Assistant (IPPA), which supports knowledge-assisted planning of machining operations and pre-sented an opportunity for CAD/CAM integration Inthe same year, Jiang et al (1999) created an automaticprocess planning system (APPS) for the generation ofmanufacturing process plans directly from CADdrawings The APPS uses knowledge such as machinelimitations, tooling availability and other process-related manufacturing information

In CAPP, selection of cutting tools and tion of machining conditions require a considerableamount of experience and knowledge The objectivesmay include selecting the best tool holders and insertsfrom an available cutting tool stock, and determiningthe optimum cutting conditions Arezoo et al (2000)developed a knowledge-based system EXCATS (expertcomputer aided cutting tool selection) for selection ofcutting tools (including tool holders and inserts) andconditions of turning operations such as feed, speedand depth of cut The system demonstrates the key role

determina-of a knowledge-based system in achieving maximumflexibility in process planning automation Zhao et al.(2002) further extended EXCATS by integrating it

with a CAD system Their system is capable of

processing CAD data and automatically generating

the component representation file for EXCATS Pham

and Gologlu (2001) designed a hybrid CAPP system

called ProPlanner, to facilitate concurrent product

development In their work, a hybrid knowledge

representation scheme, and objects were used to store

domain-related declarative knowledge and production

rules used to codify procedural knowledge Gologlu

(2004) extended the ProPlanner, and presented an

efficient heuristic algorithm for finding near-optimal

operation sequences from all available process plans in

a machining set-up In the adopted approach, a

four-level hierarchy was used: feature four-level, machining

scheme level, operation level and tool level This

enabled the problem of operation sequencing to be

systematically addressed

3.3 Neural networks

Neural networks are the techniques developed by

simulating the human neuron function and using

weights distributed among their neurons to perform

implicit inferences (Ming et al 1999) The function of a

neural network-based system is determined by four

parameters: net topology, training or learning rules,

input node characteristics and output node

character-istics (Prabhakar and Henderson 1992) Use of neural

networks can give a process planning system an

adaptive and learning capability Neural network has

several advantages over other methods used in CAPP

(Yue et al 2002) It can tolerate slight errors from

input It is usually faster because the process is limited

to simple mathematical computations and does not use

either a search or a logical rule to parse information A

neural network also has the ability to derive rules or

knowledge through training with examples and can

allow exceptions and irregularities in the knowledge/

rule base Using a neural network, one can easily

consider multiple constraints in parallel

The pioneering research on a neural network

approach using perception for feature recognition was

proposed by Hwang and Henderson (1992) Perception

is a pattern classifier for only linearly separable patterns,

with supervised training The network, trained to

recognise intermediate features such as a pocket, a slot

and a through-hole, was able to recognise partial

features presented but failed to recognise more complex

features, such as a cross-slot In the article by Onwubolu

(1999), a back propagation neural network was applied

to the problem of feature recognition Devireddy and

Ghosh (1999) presented a methodology of integrating

design with process planning using neural networks

The system can be trained to handle new types of

components

Because it is easy to modularise a neural networksolution for a particular problem, neural networks arefound to be combined with a number of other methods

A hybrid intelligent inference model for CAPP wasdeveloped by Ming et al (1999) Their model combinesthe advantages of both expert systems and neuralnetworks The methodology provided an effective means

to administrate, control and coordinate the CAPPfunctions It also enhanced the adaptability andflexibility of the CAPP system to cope with the dynamicnature of the manufacturing environment Chang andChang (2000) developed a system that integrates thevariant and generative forms of CAPP It consists ofprocess planning expert system modules and a dynamiclearning recognition mechanism, through integration offuzzy logic rules, artificial neural networks and expertsystems It is able to decide whether to use the variant orgenerative procedures Fuzzy logic rules and neuralnetworks enable process planning to have a dynamicadaptive learning ability Ben Yahia et al (2002)presented a feed-forward neural network based systemfor CAPP The methodology can cater for some difficultproblems Ming and Mak (2000a) formulated theproblem of selecting exactly one representative from aset of alternative process plans for each part, tominimise, for all the parts to be manufactured, the sum

of both the costs of the selected process plans and thedissimilarities in their manufacturing resource require-ments They combined a Hopfield neural network(Hopfield and Tank 1985) and GAs to solve the aboveproblem Later, the same authors used the Hopfieldneural network to solve the manufacturing operationselection problem (Ming and Mak 2001) Similarly, Deb

et al (2006) used the back-propagation neural networkmethod for the selection of all possible operations formachining rotationally symmetrical components Thiswas done by pre-structuring the neural network withprior domain knowledge in the form of heuristic orthumb rules Ding et al (2005) presented an optimisa-tion strategy for process sequencing based on multi-objective fitness: minimum manufacturing cost, shortestmanufacturing time and best satisfaction of manufac-turing sequence rules They used an artificial neuralnetwork to allocate the relative weights for the threemain evaluating factors for process sequencing, andapplied an analytical hierarchical process to evaluate thesatisfaction degree of the manufacturing sequence rulesfor process sequencing Amaitik and Kilic (2007)developed a process planning system for prismatic partswhere several neural network models were developed.The main neural network model is utilised to selectproper cutting tool(s) for each machining feature Theidea is that for each machining feature and machiningoperation combination there is a corresponding cuttingtool to be used to create that feature The neural network

is trained based on this criterion For each cutting tool, a

neural network was designed and trained to select the

proper tool geometry Selection of a machine tool on

which the machining operations can be performed to

produce the given part is also implemented by a neural

network The input vector of the neural network

includes machining part characteristics and machining

operation characteristics, and the output vector of the

neural network contains recommended specifications of

the machine tool to be used to perform the task These

recommended specifications are used to search in an

available machine tool database for a proper machine

tool Sunil and Pande (2008) proposed a 12-node vector

scheme to represent machining feature families having

variations in topology and geometry The data of the

recognised features was then post-processed and linked

to a feature-based CAPP system for CNC machining

To solve setup planning problems, Ming and Mak

(2000b) used Kohonen self-organising neural networks

and Hopfield neural networks Kohonen self-organising

neural networks were specially developed to solve the

setup generation problem, by considering fixtures/jigs

constraint, approach direction constraint, feature

pre-cedence relationship constraint, and position tolerance

relationship constraint Hopfield neural network was

adopted to solve the operation sequence problem and

the setup sequence problem These two problems are

NP-complete problems, and can be mapped onto the

travelling salesman problem by numerating the

con-straints, in the operation sequence and the setup

sequence, into the distance among operations and

among setups

Park et al (2000, 2001) developed a methodology

for incremental supervised learning of cutting

condi-tions for process planning It enabled the model for

generating cutting conditions to be enhanced while the

system is in continual use The methodology of the

fuzzy neural network is applied to model the process of

learning and enhancing cutting condition, where it is

capable of online and offline supervised learning in

response to arbitrary sequences of analogue and binary

input / target vector pairs The issue of intelligent

tool-path generation was addressed by Balic and Korosec

(2002) They presented a discussion to show that

artificial neural network is able to establish a desirable

milling tool-path strategy/sequence for free-surface

machining Joo et al (2001) presented a dynamic

planning model for determining cutting parameters

Using neural networks, they developed a dynamic

planning model for determining cutting parameters

and this model is executed by a shop-floor controller to

determine the cutting parameters for the removal

feature based on current shop-floor status

Devireddy et al (2002) proposed a three-layer,

back-propagation neural network for selection of

machining operations for all the features at a time,

by taking into consideration the global sequencing ofoperations across all the features of a part Thisapproach is able to overcome some limitations ofdecision trees and expert system-based approaches.Korosec et al (2005) reported a neural-fuzzy modelthat uses the concept of ‘feature manufacturability’ toidentify and recognise the degree of difficulty inmachining The model was created by means ofconstructing parametric fuzzy membership functions,based on the neural networks learning process Athree-layer, feed-forward architecture was used Thesystem created was successfully implemented by Balicand Korosec (2002) for intelligent tool-path generationfor free-form surface machining

3.4 Genetic algorithms

A GA is an intelligent search method requiringdomain-specific knowledge to solve a problem It hasbeen successfully applied to various optimisationproblems since the mid-1960s GAs are in the category

of post-collation optimisation approach (Qiao et al.2000) By mimicking the evolutionary process ofnature, such algorithms have been employed as globalsearch and optimisation techniques for various scien-tific and engineering problems GAs search from apopulation of points, unlike the enumerative techni-ques where the objective function is calculated at eachpoint in a search space, one point at a time GAs mimicthe process of natural evolution by combining thesurvival of the fittest among solution structures with astructured, yet randomised, information exchange andoffspring creation The offspring displaces weak solu-tions during each generation Therefore, GAs are verysimple, straight forward, yet powerful methods forglobal search and optimisation of multimodal func-tions (Singh et al 2003)

The main advantage that the GA-based approachhas over other CAPP approaches is in the task ofconcurrently considering machine tools, cutting tools,tool access directions for each operation, and thesequence among the operations Therefore, the result-ing process plan model retains the entire solutionspace This makes it possible to find a globally optimalprocess plan for a part The CAPP model for machinedparts to be made in a job shop environment as reported

in Zhang et al (1997) is such an example Similar workhas been done by Rocha et al (1999), with the use of

GA in a CAPP system to generate the sequence ofoperations and to select machine tools and cuttingtools that minimise machining time In the systemdeveloped by Dereli and Filiz (1999), a reward/penaltymatrix called REPMAX for each setup was determinedbased on the selected criterion, such as safety or

minimum tool change The objective of optimisation

was to gain the least total penalty or largest total

reward Bo et al (2006) reconstructed GAs based on

the analysis of various constraints in process route

sequencing, including the establishment of coding

strategy, evaluation operators and fitness function, to

meet the requirement of sequencing work The

men-tioned constraints are used as the control strategy for

GAs in the searching process to direct the calculation

of GAs, and to find the optimal result that can satisfy

the constraints Li et al (2000a) used GAs to optimize

machining datum selection and machining tolerance

allocation in process planning

Much like neural network, GA is also found to be

combined with other methods to solve optimisation

problems in CAPP The use of a GA in conjunction

with a neural network is mentioned earlier in the work

of Ming and Mak (2000b) The weakness of the

Hopfield neural network that leads to a local optimal

solution to the problem was eliminated by the use of

GA Genetic operations, such as individual evaluation,

parent selection, reproduction and mutation, are

performed to obtain the best individual as the solution

Therefore, the combination approach of the hybrid

Hopfield network is able to obtain the near-global

optimal solution to the process plan selection problem

The combined approach is also adopted by Ding et al

(2005) They presented an optimisation strategy for

process sequencing based on multi-objective fitness:

minimum manufacturing cost, shortest manufacturing

time and best satisfaction of manufacturing sequence

rules They proposed to incorporate the GA, neural

network and analytical hierarchical process (AHP) for

process sequencing A globally optimised fitness

function is then defined including the evaluation of

manufacturing rules using AHP, calculation of cost

and time, and determination of relative weights using

neural network techniques

Li et al (2002) developed a hybrid GA-SA

(simulated annealing) approach to solve the

optimisa-tion problem of process planning for prismatic parts

In this approach, the assignment of machining

resources, selection of set-up plans and sequencing of

machining operations are considered concurrently The

advantage of this hybrid GA-SA approach is that it

can generate multiple optimal or near-optimal process

plans, with acceptable computation efficiency based on

a combined machining cost criterion with weights

Based on the multiple process plans, process planners

can make a more accurate and flexible decision

according to the actual conditions This approach

can conveniently simulate a practical and dynamic

workshop environment, considering the unavailability

of a machine or tool in bottleneck (competition) usage

or breakdown, change of machining cost evaluation

strategy, and substitution of machines or tools inanother shop floor Qin et al (2005) introduced a fuzzyinference system for choosing appropriate machines

In addition, the load for each machine is balanced byusing the GA based on the capability information,which is measured by a reliability index For the mostreliable machine, the load given will be more than theunreliable one The load on each machine is measured

by the machine utilisation, i.e the percentage of timethe machine is being utilised

Bhaskara Reddy et al (1999) proposed a quickidentification of (near) optimal operation sequences in

a dynamic planning environment using a GA Theyidentified the feasible sequences based on a FeaturePrecedence Graph and used minimum production cost

as the objective function A precedence cost matrix wasgenerated for any pair of features based on the relativecosts corresponding to the number of tasks that needed

to be performed In the following year, Qiao et al.(2000) proposed a GA-based operation sequencingmethod that provides a potential for finding ‘good’machining operation sequences A fitness function isdeveloped for this purpose, considering multipleprocess planning rules simultaneously and flexibly.The value of the fitness function is a criterion toevaluate the degree of satisfaction of a searchedoperation sequence to generate a feasible operationsequence and, eventually, a near-optimal solution

In the work of Shunmugam et al (2000, 2002), theyconsidered a face-milling operation for selection ofmachining parameters such as number of passes, depth

of cut in each pass, speed and feed, which wereobtained using a GA, to yield minimum total produc-tion cost while considering technological constraintssuch as allowable speed and feed, dimensionalaccuracy, surface finish, tool wear and machine toolcapabilities From experiments, the method proposedalways yields less production cost than, or equal to,that by other methods

Chiung et al (1998) proposed a GA-based tionary approach to solve the sequencing problem bysimultaneously considering the operation flexibility,realistic shop factors and transportation time of anAGV (automated-guided vehicle) system It wasformulated as a bi-criteria mathematical model mini-mising the total processing and transportation timeand minimising the load variation between machines

evolu-Li et al (2005) presented a GA to search for an optimalprocess plan for a single manufacturing system as well

as distributed manufacturing systems according toprescribed criteria such as minimising processing time

By applying a GA, their CAPP system could generateoptimal or near-optimal process plans based on thechosen criterion They claimed that the algorithmadopted a crossover operator described by Bhaskara

Reddy et al (1999) The developed technique is

comparative or better in dealing with process planning

in a single manufacturing system or factory Salehi and

Tavakkoli-Moghaddam (2009) applied GAs for

pro-cess planning in both the preliminary and detailed

planning stages In the preliminary stage, feasible

sequences of operations are generated based on the

analysis of constraints In the detailed planning stage,

the GA prunes the initial feasible sequences to give an

optimised operation sequence and optimised selection

of machine, cutting tool, and tool approach direction

for each operation

3.5 Fuzzy set theory/logic

Much of the decision-making in the real world such as

process planning takes place in an environment in

which goals and constraints are fuzzy, i.e not known

precisely This generates the need of approximation to

obtain a reasonable model of a real system Fuzzy

theory deals with this type of problems by

transform-ing human knowledge to mathematical formulae (Beg

and Shunmugam 2003) and puts it into engineering

systems together with other information like

mathe-matical models and sensory measurements (Wang

1997), where goals and constraints can be modelled

by fuzzy sets

Xu and Hinduja (1997) developed a fuzzy decision

system to evaluate the tolerances in a design model for

a decision whether only a roughing operation will

suffice or a semi-finishing and/or finishing operation

are needed Fuzzy logic is often combined with other

methodologies to solve process-planning problems

Chang and Chang (2000) proposed a system that

integrates the variant and generative CAPP It consists

of process planning expert system modules and a

dynamic learning recognition mechanism Fuzzy logic

rules, artificial neural networks and expert systems are

used Fuzzy logic and neural network techniques are

used for dynamic and adaptive learning, where fuzzy

set theory provides a suitable tool to deal with

uncertainty or ambiguity problems Wong et al

(2003) created a prototype process planning system

that uses a hybrid of fuzzy and genetic approaches for

solving the process-sequencing problem The

cost-tolerance relationship is developed by fuzzy linguistic

variables and fuzzy ‘if-then’ rules are established The

imprecise manufacturing information, such as tool

setup cost, is expressed in fuzzy numbers Zhao et al

(2004) introduced a fuzzy inference system for the

purpose of choosing appropriate machines, as an

alternative way to integrate the production capability

during scheduling In addition, based on the capability

information, the load for each machine is balanced

by using a GA This is used to overcome the problem

that if a machine is unreliable it is not being utilised

at all

Park et al (2000) presented a methodology thatenabled the model of generating cutting conditions to beenhanced while the system was in continual use Wang

et al (2001) developed a method that used greyrelational analysis and fuzzy clustering to formpart families efficiently, based on factors such asprocessing time, lot size, and operation sequence Ingrey relational analysis, black represents having noinformation and white represents having all informa-tion A grey system has a level of information betweenblack and white In the developed system, grey relationalanalysis is used to relate important product processfactors and obtain a similarity matrix, and fuzzyclustering is used to form part families by truncatingthe transitive closure of the similarity matrix

3.6 Petri netsNowadays, process planning is required to be flexibleenough to meet the requirements in dynamic manufac-turing Recent research on alternative process planningrepresents another trend (Wu et al 2002) PNs have theability to represent and analyse concurrency andsynchronisation phenomena in an easy way, such asconcurrent evolutions, where various processes thatevolve simultaneously are partially independent (Li et al.2000b) Furthermore, PN approach can be easilycombined with other techniques and theories such asobject-oriented programming, fuzzy theory, neural net-works, etc These modified PNs have also foundapplications in process planning PNs have an inherentquality in representing logic in an intuitive and visualway Based on the basic PNs, fuzzy PNs (FPNs) havebeen developed to address issues characterised byuncertainty, imprecision and ambiguity

Kiritsis et al (1999) considered cost estimation ofoperation sequencing in nonlinear process planning,i.e taking into consideration of processing alterna-tives To determine overall costs for a feasible processplan, they took into account the costs caused bymachine, setup and tool changing in addition to thepure operation cost They developed two PN techni-ques for process planning cost estimation; both arebased on a new PN model (PP-net: Process Planningnet) that allows the modelling of partial process plans.The first technique is based on building a complex PNcalled PPC-system (Process Planning Cost system) byintegrating the PP-net and separate PNs describing thecosts of machine, setup and tool changing The secondmethod proceeds with cost calculation by attaching aspecific data structure to each PP-net transition thatdescribes the associated machine, setup and tool forthe operation modelled by that transition

Because the knowledge in an expert system is often

vague and constantly updated, the expert systems are

intrinsically fuzzy and dynamic systems It is important

to design a dynamic knowledge inference framework

that is adjustable according to such knowledge variation

as human cognition and thinking Wu et al (2002)

proposed an ordered PN (OPN) model to address

alternative and optimal operation planning with

man-ufacturing resource constraints OPN is a modified fuzzy

PN that can be utilised to represent operation planning

knowledge, where operation selection can be carried out

by solving algebraic equations to obtain the T invariant

of an OPN The state explosion problem in traditional

expert systems is then avoided Based on T invariants of

an OPN, alternative operation plans can be easily

obtained, which can act as a basis for further generation

of alternative process plans Moreover, each alternative

operation plan is assigned a numerical value that reflects

its confidence degree Li et al (2000b) proposed a

generalised FPN model, called adaptive fuzzy PN

(AFPN) AFPN not only takes the descriptive

advan-tages of FPN, but also has learning ability as neural

networks do In addition to knowledge representation

and reasoning as other FPN models do, AFPN has one

additional advantage – it is suitable for dynamic

knowledge, meaning that the weights of AFPN are

adjustable Based on the AFPN transition firing rule, a

modified back propagation learning algorithm has been

developed to assure the convergence of the weights

Canales et al (2006) presented a methodology to extend

fuzzy number approach to triangular function learning,

and introduced an adaptive fuzzy PN (AFPNT) It has

learning ability via neural networks, so that fuzzy

knowledge in a knowledge base could be learned

through an AFPNT model The fuzzy production rules

in the rule-based system are modelled by AFPNT

Similar to other FPNs, AFPNT can be used for

knowledge representation and reasoning Kasirolvalad

et al (2004) presented an AND/OR net approach for

planning machining operations, and showed that PNs

could be used to model all CNC machining operations in

a graphical manner It introduced a technique based on

AFPN (Li et al 2000b) to model, monitor and control

the surface roughness and machining operation quality

Machine tool vibration, cutting force, spindle speed,

feed-rate and machining time were used to gain high

quality surface roughness and machining operation

quality

3.7 Agent-based technology

Global competition and rapidly changing customer

requirements have led to major changes in production

styles and manufacturing strategies Traditional,

cen-tralised and sequential process planning mechanisms

are found insufficient to respond to the changingproduction styles and high-mix, low-volume produc-tion environments Agent technology offers a possibletool to address such problems, and to design andimplement efficient distributed manufacturing systems(Shen et al 2006a,b)

General issues related to agent-based ing systems have been discussed by Shen and Norrie(1999) They included agent encapsulation, multi-agentorganisation, dynamic system reconfiguration, learn-ing in agent-based manufacturing systems, design andmanufacturability assessments, distributed planning,and scheduling Most of these issues also exist in agent-based process planning systems The key issues indeveloping an agent-based process planning system areidentified as:

manufactur- Architecture issues including agent architectureand system architecture of an agent-basedprocess planning system;

Communication issues including standards andprotocols and

Application issues in support of collaborativeprocess planning

It is considered hardly feasible to develop anextensive industrial CAPP system by using only onelarge expert system Zhao and Wu (1999) and Zhao

et al (2000) proposed a system called CoCAPP, todistribute complex process-planning activities to multi-ple, specialised problem solvers, and to coordinatethem to solve complex problems CoCAPP aims tosatisfy five major requirements: autonomy, flexibility,interoperability, modularity and scalability It tacklesprocess-planning problems by distributing them tospecial process planning agents All defined process-planning agents coordinate and cooperate with eachother via a commonly shared language with theexpectation of reaching agreements when conflictsoccur These agents have their own tasks to perform,which are notified by a blackboard agent Zhang et al.(1999a,b) proposed an agent-based adaptive processplanning (AAPP) system, on top of an object-orientedmanufacturing resources modelling framework Fiveagents with distributed process knowledge were used inthe AAPP for part information classification, manu-facturing resources mapping, process planning, andmachining parameter retrieval The coordinationbetween agents was based on the contract net protocol.Chan et al (2001) developed an integrated, distributedand cooperative process planning system The processplanning tasks are broken into three levels: initialplanning, decision-making and detail planning Theydealt with manufacturability evaluation and genera-tion of alternative processing routes for parts, retrieval

and ranking of alternative process, tasks of final

selection of machines and cutting parameters, and

calculation of machining cost and time

Agent-based applications provide a new way of

viewing problems and designing solutions They are

robust and dynamic Wang and Shen (2003) presented

a DPP methodology by integrating feature-based

process planning, function block (FB)-based control

(see a sub-section on ‘Emerging technologies’ later in

this section), and agent-based decision-making The

proposed methodology is suitable for dynamic,

recon-figurable and distributed manufacturing environments

The system can deal with behaviour among a group of

autonomous agents about how they can coordinate

their goals, plans, skills, and activities to work

cooperatively toward a single global objective Depince

et al (2001) introduced a Human Integrated CAPP

system, based on a multi-agent approach They

detailed the data agent family of agent modules The

number and coherence of the machining process

components are generated using constraint agent and

optimisation agent Allen et al (2005) presented a

STEP-NC compliant computational environment used

to demonstrate agent-based process planning The

system comprises a multi-agent framework, where

agents represent the individual STEP-NC features of

the component and work independently and

coopera-tively to generate process plans Nassehi et al (2006)

examined the application of collaborative multi-agent

systems in designing an object-oriented process

plan-ning system called Multi-Agent System for CAPP, for

prismatic components in a STEP-NC compliant

environment Also using the STEP-NC information,

Fichtner et al (2006) introduces a combination of

agent-based organisation and self-learning of

feature-based technological information for acquisition and

preparation of distributed NC information in support

of NC planning More recently, Agrawal et al (2009)

presented a multi-agent system for DPP The proposed

architecture consists of three autonomous agents

(global manager agent, design agent, and optimization

agent) These agents are capable of communicating to

each other through XML

3.8 Internet-based technology

Significant changes to the enterprise strategy and

manufacturing paradigms have led to the development

of Internet/Web-based process planning and

manufac-turing to support a networked manufacmanufac-turing

envi-ronment A CAPP system must accommodate the

variation and distribution of manufacturing resources

and processing objects, cooperatively A cooperative

CAPP system mainly studies how to support

coopera-tive process planning among engineers at different

places, and how to improve instantaneous nication among them In the work by Xu et al (2005b),they put forward an idea that used screen-sharingtechnology to support cooperation This approachovercomes the limitation on process resources andknowledge in the traditional narrow-sense processplanning, and improves engineers’ cooperative work.You and Lin (2005) adopted the Java language totransfer a CAPP system to a Web-based environmentand thus distribute the operations of CAPP to variouscomputer systems to reduce the computation loading.The distributed computing environment is based onJ2EE, enabling the manufacturing processes to beplanned effectively over a network

commu-Zhang (2002) presented a research article consisting

of system architecture, main function modules, nological knowledge representation and knowledgeacquisition mechanism, reasoning mechanism adopt-ing recursive algorithm, and collaborative works in theprocess planning The developed CAPP systemadopted knowledge-based methods, written in JSPand Java language The CAPP system can receive andprocess CAD data and produce final process plans thatmeet production requirements Liu et al (2004)developed a CAPP system based on the COMcomponent technology, NET technology, and theXML technology The computing model was a mix

tech-of C/S (client/server) and B/S (browser/server) models.The basic CAPP functions were encapsulated intocomponents Customisation on such an integrationsystem can solve the problem between a general CAPPsystem and a custom CAPP system During processplanning, teams can collaborate with each other

In the research by Qiu et al (2001), a distributedmulti-user environment on the Internet was suggested

It was implemented by a combination of an externalauthoring interface (EAI) and Java Using the Web-based system, they carried out manufacturabilityevaluation based on a pre-defined process plan Inthe same year, Sun et al (2001) presented an agent-based concurrent engineering system concerning pro-duct design and manufacturing planning Liu and Peng(2005) developed an Internet-enabled system for setupplanning in machining operations using Java and Webtechnologies XML was used to transfer informationbetween various manufacturing systems Similar workwas also carried out by Peng et al (2005) Theypresented an Internet-based integrated system forsetup planning The system communicated withprocess planning, fixture design and NC programmingsystems To generate a global optimal setup plan, theirmethodology considers machining precedence toler-ance requirement and fixturing constraints Chung andPeng (2004) developed a Web-based tools and machineselection system (WTMSS) that provides intelligent

decision-making, and sharing of production

knowl-edge through the Internet Alvares et al (2008)

reported an integrated Web-based CAD/CAPP/CAM

system for the remote design and manufacture of

feature-based cylindrical parts The information about

features is manipulated through a relational database

management system A graphic user interface (GUI) is

implemented in Java and HTML Through this GUI, a

user inputs the information about design features

Then these data are sent to the server Because the part

is cylindrical, the user models the part in two

dimensions, and it can be visualised as

three-dimen-sional through VRML Hu et al (2008) carried out an

XML-based implementation of manufacturing route

sheet documents for context-sensitive and Web-based

process planning They used a

psycho-clonal-algo-rithm-based approach to solve the

operation-sequen-cing problem Agrawal et al (2009) presented a

multi-agent system for DPP The autonomous multi-agents, (i.e

global manager agent, design agent and optimization

agent) are capable of communicating to each other

through XML Hence, DPP problem is enabled in the

e-manufacturing environment

3.9 STEP-compliant CAPP

Although computers provide greater efficiency in

design and manufacturing, researchers still face the

challenge of sharing information between different

applications including CAPP, due to the

incompat-ibility of product data representations To overcome

this problem, the International Organisation for

Standardisation (ISO) initiated an effort to develop

neutral data models in support of exchanging and

sharing product information; it is called the Standard

for Exchange of Product Model Data (ISO 10303 or

STEP 1994) ISO 10303-203 (1994) is the first and

perhaps the most successful Application Protocol (AP)

developed for exchanging design data between

differ-ent CAD systems Researchers have since attempted to

develop algorithms to recognise features from a CAD

model based on AP203 AP 214 (2001) is a more

targeted application protocol; it defines the core data

for automotive mechanical design processes Most of

the contemporary CAD/CAM systems can input and

output these two types of STEP files Another

important application protocol is AP 224 (1999) for

mechanical product definitions for process planning

using machining features STEP-compliant CAPP

research has been making much inroads recently since

ISO 14649-1, 14649-10, 14649-11 (2004) and ISO

10303 AP 238 (2007) were published in 2004 and

2007, respectively They are collectively known as

STEP-NC for supporting STEP-compliant,

feature-based CNC machining

Kang et al (2003) proposed an approach tointerlink design and process planning using anintegrated product model based on the STEP format

It deals with recognition of machining features,incorporation of manufacturing information such assurface roughness and tolerances, and implementation

of a neutral interface Ong et al (2003) proposed amanufacturing feature recogniser for the Unigraphicssystem It integrates feature recognition with design-by-feature approaches to generate a STEP-basedmanufacturing feature model The design featuremodels and feature recognition system are integratedinto a concurrent engineering agent environment ASTEP-based feature modeller (STEP-FM) has beendeveloped by Amaitik and Kilic (2005) as a design-by-feature tool to integrate design and manufacturingtasks STEP-FM uses high-level 3D features as thebasic design entities in the design process Designerscan also consider manufacturing properties earlier inthe part design phase The part data file, which is saved

in the STEP XML format, can be passed directly to thedownstream CAPP activities without using featurerecognition process Lau et al (2005) worked onfeature recognition for automatic process planningwhere STEP design files are used as the informationsource for generating detailed manufacturing process.Yifei et al (2008) presented an automatic features-extraction and process-planning system using theSTEP AP214 data format In their system, a methodfor extracting the basic features from a STEP AP214 of3D model was performed after analysing the STEPAP214 file The combination rules of basic featureswere then studied by identifying feature relationships.Also, an approach to processing tolerance information

is proposed by utilising the CAD system’s provided functions for macro recording and editing.Finally, a FEB process-planning system prototype wasdeveloped employing the knowledge-based approach

self-In the article by Rameshbabu and Shunmugam (2009),

a hybrid approach that uses volume subtraction andface adjacency graph was proposed to recognisemanufacturing features from 3-D model data inSTEP AP-203 format The recognised manufacturingfeatures were then clustered based on preferential basefor machining Setup sequences are obtained byalternative rating and ranking

In their Java-based process-planning proposal,You and Lin (2005) used STEP AP224 to define themanufacturing features By using STEP AP224 datamodel to bridge CAD and CAPP systems, therelationships among product, shape and featuredefinitions are maintained STEP is also incorporated

in the system created by Peng et al (2005) The inputuses a format file of STEP which increases the utilityand applicability of the system Amaitik and Kilic

(2007) presented a process planning system using

STEP features (ST-FeatCAPP) for prismatic parts

The system maps a STEP AP224 XML data file,

without using a complex feature recognition process,

and produces the corresponding machining operations

to generate the process plan and corresponding

STEP-NC in XML format One of the main purposes of this

CAPP system is to integrate a standardised

feature-based model with process planning utilising the

concept of STEP-based features In the GF-CAPP

system developed by Gonzalez and Rosado (2003), AP

224 features are also used

STEP-compliant process planning research has

been reported in a number of publications since the

STEP-NC was initiated and published Discerning

readers are referred to the review articles (Xu and He

2004, Xu et al 2005a, Zhao et al 2009), which report

on the research work prior to and post 2004,

respectively The remaining of this paragraph

high-lights some of the recent work Nassehi et al (2006)

examined the application of distributed artificial

intelligence methods, namely collaborative multi-agent

systems in designing an object-oriented process

plan-ning system for prismatic components in a STEP-NC

compliant environment In Chung and Suh’s work

(2008), the proposition of a nonlinear process-planning

based on the STEP-NC paradigm was presented and

an optimal solution algorithm for this type of process

planning was developed The algorithm is based on a

branch-and-bound approach and heuristics derived

from engineering insights The developed process

planning method and optimisation algorithm were

implemented and tested via a system called TurnSTEP

The article by Nassehi et al (2007) introduces a

software platform entitled the Integrated Platform for

Process Planning and Control (IP3AC) to support

STEP-NC compliant process planning A prototype

process planning system (PPS) based on the platform

was also presented as a sample application in the light

of future interoperable planning and manufacture

Stroud and Xirouchakis (2006) extended the feature

definitions in STEP-NC to support process planning

for aesthetic products in stone manufacturing industry

such as dry high-speed milling of marble and industrial

ceramics The new aesthetic features are divided into

four categories: shell_feature, rebate_feature,

runof-f_area and pattern_feature Machining strategies for

these new features were also suggested Fichtner et al

(2006) also used STEP-NC information for acquisition

and preparation of distributed NC information in

support of NC planning Because of the integrated

nature of STEP and STEP-NC standards, it is possible

to address process planning in a broad sense One of

the examples is the prototype system for closed-loop

machining processes, which includes generation and

execution of a STEP-NC program and feedback ofmeasured results (Brecher et al 2006) Recently, Yusof

et al (2009) described a STEP-compliant CAD/CAPP/CAM system for the manufacture of asymmetric parts

on CNC turn/mill machine A structured view of aSTEP-compliant system framework for turn/millcomponent manufacture was also provided

Considering an ultimate scenario with STEP, Zhao

et al (2009) painted an integrated STEP-compliantmanufacturing system as shown in Figure 1 Thegeometric representation data described in AP203 orother formats are translated into machining featuresdefined in AP224 The machining feature definitionsare used as inputs to macro process planning applica-tions (e.g AP240 for machining, AP223 for casting,and AP229 for forging) Micro process planning formachining (AP238) and inspection (AP219) are thencarried out for each of the aforementioned applicationprocesses In such a system, the need for dataconversion is eliminated

3.10 Emerging technologies

In recent years, some emerging technologies have beeneither developed or adapted for supporting processplanning tasks Utilisation of FBs in CAPP proposed

by Wang et al (2003, 2007) is one of them In thisresearch, the tasks of process planning are divided intotwo groups and accomplished at two different levels:shop-level supervisory planning and controller-leveloperation planning, as shown in Figure 2 Event-driven FBs are used to encapsulate high-level genericprocess plans and low-level planning algorithms Theyare then dispatched to available machines for machine-specific operation planning by triggering appropriateembedded algorithms of the FBs Four types of basicFBs are defined: machining feature FB, event switch

FB, comm\unication FB, and management FB (Wang

et al 2008, 2009) They can generate detailed andadaptive operation plans at runtime to best utilise thecapability of the available machines, and enable real-time execution monitoring of the FBs

Recognising the combinatorial nature and complexprecedence relations in process planning, Dashora

et al (2008) proposed a psycho-clonal-algorithm-basedapproach to solve optimally the operation-sequencingproblem The objective function is made morecomprehensive for the part types of varying complex-ities This approach is an extension of the artificialimmune system approach and inherits its character-istics from the Maslow’s need hierarchy theory related

to psychology The various need levels present in thealgorithm help in maintaining the viability of solution,whereas the path towards optima is revealed by thetrait of affinity maturation Hu et al (2008) used the

Figure 1 An integrated STEP-compliant manufacturing system.

Figure 2 Function Blocks in distributed process planning

same method to solve the operation sequencing

problem for Web-based process planning

To discover the typical process route in the process

planning database, Liu et al (2007) applied

Knowl-edge Discovery in Database (KDD) process planning

Process data selection, process data purge and process

data transformation are employed to get optimised

process data Cluster analysis is adopted as the

algorithm mining the typical process route A

mathe-matical model describing the process route was built by

the data matrix There are three similarities in process

route clustering: the similarity between operations was

measured by the Manhattan distance based on

operation code; the similarity between process routes

was calculated by the Euclidean distance and expressed

as a dissimilarity matrix; the similarity between process

route clusters was evaluated by the average distance

based on the dissimilarity matrix Then, the process

route clusters were eventually merged by the

agglom-erative hierarchical clustering method The process

routes clustering result was determined by the

cluster-ing granularity of process route

4 Some topical areas of CAPP

This section is devoted to some typical issues in process

planning, i.e tool selection, operation selection and

sequencing, decision models in CAPP, integration of

process planning with a production system, non-linear

process planning, energy-conscious and

energy-effi-cient CAPP, and commercialisation of CAPP research

Some of these topical areas are in fact critical tasks

in arriving at a suitable process plan; others are

important aspects of process planning research

4.1 Tool selection

Tool selection is perhaps one of the most important

functions in a process planning system This is because

the selection of a tool affects determination of

machining parameters, selection of jigs and fixture,

production rate, cost of the product, and quality

Traditionally, tool selection was manually done by

experienced operators and engineers, but numerous

errors and inconsistencies often occurred In general,

the selection of a wrong tool will lead to infeasible and

inconsistent process plans, and in turn lead to

discarding process plans Thus, there is a need for a

systematic approach to selecting ideal tools In the

research of Fernandes and Raja (2000), they developed

a systematic method for selecting the best tool set for a

given part It used an object-oriented methodology for

selection of the tooling parameters The system was

able to be incorporated into both static and dynamic

process planning systems You et al (2007) addressed

the issue of choosing the optimal cutting toolcombination for machining pockets, through slicing a3D part into 2D pockets for generating tool paths.They developed a method in which an upper boundO(N) was used to choose the optimal cutting toolcombination, where N is the number of feasible toolsavailable Meseguer and Gonzalez (2008) developed amethodology for cutting tool management based onthe use of alternative tools In this methodology, allpossible tool alternatives for each operation are singledout The tool manager sought the interferencesbetween the alternative tools keeping scheduling inmind Interferences occur when the same tools areneeded by a number of machining operations

4.2 Setup planningSetup planning is another integral part of CAPP Itinvolves preparing detailed work instructions forsetting up a part, and ensuring the precision of themachining processes and the finished component.Zhang and Lin (1999a) introduced a hybrid graphapproach for setup planning Huang and Xu (2003)presented an integrated methodology for generation ofsetup plans and selection of setup datum to determine

a setup sequence The methodology uses the and-conquer’ principle, where the complex setupplanning problem is divided into simpler sub-pro-blems: geometry analysis, precedence constraint ana-lysis, tolerance analysis, kinematics analysis and forceanalysis Frank et al (2006) studied the visibility ofslice geometry (orthogonal to the axis of rotation) todetermine setup orientations Joneja and Chang (1999)considered machining sequence constraints, machinetools and the feasibility of fixturing when carrying outsetup planning

‘divide-It is generally believed that tolerances are a majorfactor in making decisions about setups In the work ofKrishna and Rao (2006), they presented a procedure tosimultaneously allocate both design and manufactur-ing tolerances based on the minimum total manufac-turing cost Their optimisation model for tolerancedesign is a non-linear multivariable problem Theyapplied scatter search and successfully determined theoptimal tolerances at the minimum manufacturingcost To integrate tolerance for manufacturing dimen-sions in a CAPP system, Hamou et al (2006)developed a statistical and cost-based tolerance synth-esis model It is shown that this model performs anobjective and global distribution of the residualtolerances of the functional dimensions on all themanufacturing dimensions of the process plan Thedispersions method is used in the modelling process todetermine the variables of the objective function and toautomatically extract the manufacturing tolerance

chains These chains are then used to construct the

functional constraints of the optimisation model

Hebbal and Mehta (2008) developed a formalised

procedure for automatic generation of feasible setups

and selection of an optimal setup plan for machining

features of a prismatic part The proposed work

simultaneously considers the basic concepts of setup

planning from both machining and fixturing

view-points to formulate feasible setup plans To solve setup

problems for turned parts, Valin’o et al (2007)

developed a methodology that takes into consideration

constraints such as the geometry of the stock and final

part, the geometry and capacity of the chuck, and part

tolerances In general, these constraints allow the

system to obtain several valid solutions for clamping

the part Some criteria based on the clamping force

and the value of tolerances have been considered to

establish a preference order among these solutions

Finally, the analysis of linked tolerances and the tool

approach direction to each surface determine the sets

of surfaces to be machined within each set-up

Some researchers have also taken into account the

manufacturing resources during the decision-making

process The work by Yao et al (2007) is one example

A setup planning system was developed based on the

analyses of both tolerances and manufacturing resource

capability The setup planning is divided into two levels:

setup planning in single part level and in machine

station level Zhang (2008) formulated a mathematical

model to describe the tolerance information and

datum-machining feature relationship based on

ex-tended graphics The algorithm identifies the machining

features and datum, and optimises setup groups based

on the manufacturing resource capability and tolerance

analysis, as well as minimise the influence of the locating

error stack-up on the machining quality

More recently, Cai et al (2009) proposed a GA-based

adaptive setup planning approach taking into

considera-tion the availability and capacity of machine tools on a

shop floor It first generates setup plans for three-axis

machines with a single tool approach direction Because

most machine tools have three axes or more, the

three-axis based setup plans are applicable to and generic for

these machines It then considers setup merging if a

four-or five-axis machine is chosen The merged final setup

plan can best utilise the capability of the machine This

approach is integrated with a scheduling system so that

setup planning, especially the setup merging, can be done

adaptively according to the currently available machines

and their characteristics

4.3 Operations selection and sequencing

In process planning, a machining sequence is generally

obtained considering objectives such as the shortest

time and/or the minimum cost This is why a number

of researchers have considered cost as one of the majorfactors in their research Ming and Mak (2001) usedthe Hopfield neural network to solve the manufactur-ing operation selection problem Pan et al (2002)proposed a cost-estimating model in the productdesign phase A process planning based cost estimatingsystem (PPBCE) was developed based on this estimat-ing model The main modules are feature recognition,feature modelling, process planning and cost estimat-ing A system to assist selection of parameters inmilling processes was developed by Alberti et al.(2005) The algorithm minimises the cost of operations.Lee et al (2004) considered the operation selection andsequencing with the objective of minimising the costs

of operation processing, machine, setup and toolchanges The problem considered by the authors isthe same as that by Kiritsis et al (1999), in which Petri-nets were used to model and solve the problem.Although the Petri-net technique can capture systemdynamics and physical constraints, it is not adequate

to solve optimisation problems Hence, it is mented with optimisation algorithms To do this, atree-structured precedence graph was suggested torepresent alternative operations and their precedencerelations Then, using the graph, three iterativealgorithms, one optimal and two heuristics weresuggested after decomposing the entire problem intothe operation selection and operation sequence gen-eration problems Each of the iterative algorithmsembeds the shortest path algorithm for optimaloperation selection and the enumeration method togenerate all feasible sequences that satisfy the pre-cedence constraints

comple-Culler and Burd (2007) demonstrated the ture in which customer service, CAPP and a costingmethodology known as activity based costing (ABC)were incorporated into one system, thereby allowingcompanies to monitor and study how expenditures areincurred and which resources are being used by eachjob Yougtao and Jingying (2006) constructed aprocess-planning model for hole-making processes Itconsists of three parts: a features framework, aprecedent relation net and a sequencing mathematicalmodel The features framework does the mapping from

architec-a marchitec-anufarchitec-acturing fearchitec-ature (e.g hole) to its marchitec-achiningoperation(s) A semantic net named precedence-rela-tions-net reflects the precedence relationships amongthe machining operations of holes The mathematicalsequencing model generates an optimal process plan ineach operation direction by minimising the number oftool changes and decreasing the number of operationsteps Sormaz and Khoshnevis (2003) talked not onlyabout generation of alternative process plans in anintegrated manufacturing system but also described

the methods for the selection of an optimal process

plan

4.4 Decision models

Decision models, in one form or another, exist in

almost all CAPP systems Carpenter and Maropoulos

(2000) developed a decision support system for process

planning of milling operations called OPTIMUM

Unlike most of the other CAPP systems, the

machin-ability assessor in OPTIMUM allows the generation of

conservative, initial cutting data from incomplete or

fuzzy input data As the system is implemented

according to the theory of data-driven design, the

performance of the machinability assessor can be

upgraded by adding more company-specific historical

cutting data to the main data tables Vidal et al (2005)

presented a research article focused on the problem of

choosing a proper manufacturing route The aim was

to design a system to help with selection of parameters

in the cutting process for milling The algorithm

created was based on optimising the cost of machining

operations The selection of parameters takes into

account all the existing restrictive factors (material,

geometry, roughness, machine and tool)

Tu et al (2000) proposed a reference framework for

process planning for a virtual one-of-a-kind (OKP)

company The proposed method provides a useful

process planning method to deal with the continuous

customer influences through the production The

knowledge and skills of human process planning

experts were used to generate a proper process plan

for a customised product or OKP product This model

is used as an optimisation tool to select production

resources among the partner plants and finally refine

the process plan

Xu and Li (2008a,b, 2009), Xu et al (2008) and Xu

et al (2009) presented an atomic inference engine

model for process parameter selection using

mathema-tical logic The methodology of modelling the inference

mechanism of process parameter selection is proposed

with backward chaining of mathematical logic that is a

form of goal-directed reasoning To represent process

knowledge and enhance automation and flexibility of

making-decision processes, a rule-fused method for

representation of typical processes and automatic

decision was presented by Huang et al (2008) Relative

information of a typical process is divided into

invariable and variable parts The invariable part is

expressed by the united model of the typical process,

whereas the variable part is expressed by production

rules embedded into the typical process A decision

support system was developed to perform

semi-structured tasks such as setup planning and

establish-ing precedence relationship among various machinestablish-ing

operations (Kumar and Rajotia 2005) for metric components in a job shop environment

axisym-4.5 Integrated process planning and knowledgerepresentation

It is perhaps not uncommon to receive complaintsfrom a shop-floor about the generated process planthat is impractical and/or unfit for the entire produc-tion system There is therefore a need for theintegration of process planning and scheduling andeven the entire production system In the research bySaygin and Kilic (1999), an integration framework insupport of flexible process plans and offline schedulingwas developed The aims were to highlight the need for

an effective integration of process planning andscheduling to increase the potential for enhancedsystem performance and to improve decision-makingduring scheduling Min et al (2004) developed anintegrated system of CAPP (ICAPPS), aiming forsingle-piece, small-lot and make-to-order production

On the basis of the integrated model, the authorsconstructed the function models of ICAPPS, whichinclude the designing layer, part planning layer, shopplanning layer and scheduling layer Meseguer andGonzalez (2008) developed a tool management systemthat can generate a collection of tool alternativescompatible with the scheduling system Tool manage-ment can then use these sets of tools to plan toolchanges and to react to perturbations in the produc-tion system Ciurana et al (2008) developed anintegrated process planning and scheduling tool usingthe integrated definition (IDEF) methodology Anactivity model was used to develop the system thatallows the user to plan the process and the production

at the same time Knowledge integration duringproduct design has been studied (Roucoules et al.2003) This was realised through the use of twofeatures-based tools for process planning knowledgeintegration: a What-if system and a CAPP system.Because the process plans are generated on demandwith the manufacturing resources on a shop floor,modelling method of the manufacturing resources isalso an important issue in CAPP Xu and Li (2008a,b)and Xu et al (2008) created a clustering-basedmodelling scheme for manufacturing resources Thismodelling scheme combines clustering method withaveraging method so as to reasonably partition andclassify the manufacturing resources on the shop floor.The same authors proposed a meta-modelling para-digm of manufacturing resources using the first- andsecond-order logic Based on the above work, the sameresearch group built a process-planning schema (Xu

et al 2009) Sormaz and Khoshnevis (2003) discussed amethod for generating alternative process plans in a

manufacturing system that takes into account

produc-tion schedules

4.6 Nonlinear process planning

Given a machining task, any number of possible

process plans may exist depending on the actual

machine tool used Even for the same machine tool,

it is likely that there are more than one way of doing

the job Therefore, planning with alternatives sounds

like a practical approach, which is also sometimes

called nonlinear process planning Zhang et al

(1999a,b) presented a novel CAPP model for machined

parts in a job shop environment that contains

customer-specified machine tools and cutters The

approach models process planning problems in a

concurrent manner to generate the entire solution

space by considering the multiple planning tasks, i.e

operations (machine, tool and tool approach direction)

selection and operations sequencing simultaneously

Precedence relationships among all the operations

required for a given part are used as the constraints

for the solution space The relationship between an

actual sequence and the feasibility of applying an

operation is also considered An algorithm based on

simulated annealing (SA) has been developed to find

the optimal solution

Kiritsis et al (1999) considered cost estimation of

operation sequencing in nonlinear process planning,

i.e taking into consideration of processing

alterna-tives To determine overall costs for a feasible process

plan, they took into account the costs caused by

machine, setup and tool changing in addition to the

pure operation cost Based on a multi-agent

architec-ture, De´pince´ et al (2001) generated a nonlinear

process plan Jang et al (2003) presented elaboration

and validation methodologies for AND/OR

graph-based nonlinear process plans, which help users

construct, validate or modify a controller-friendly

process plan Tian et al (2007) utilised STEP-NC

data model to develop and validate nonlinear process

plans They used so-called elitist selection GAs for

optimising process routes Likewise, Chung and Suh

(2008) also presented a nonlinear process planning

method based on STEP-NC It is based on a

branch-and-bound approach and heuristics derived from

engineering insights

Nonlinear process planning is further complicated

by issues such as possible alternative feature-based

interpretations of the same part, hence different

process plans can be generated for the same part

Nonlinear process planning deserves further research

as it opens avenues for scheduling with alternative

routings, which is in fact the key issue in CAPP and

scheduling integration Shin et al (2001) and Ferreira

et al (2001) went one step further by investigating theinfluence of alternative process plans in a dynamicshop floor environment

4.7 Energy-conscious and energy-efficient CAPPToday’s paradigm shift towards environmentally con-scious production has seen a growing trend since themid 1990s to incorporate such factors in processplanning research Sheng et al (1995) in their articleoutlined an environmentally conscious multi-objectiveapproach to process planning In this approach theydefined a feed-forward model which takes into accountenvironmental factors such as process energy, processtime, fluid coated on chips, evaporated fluid, tool scrapfluid mist, chip volumes and tool particles Thisinformation is fed into an environmental impact modelwhere a score for each machining operation isgenerated and fed into a process planning modulewith process energy used and process time and surfacequality requirements to generate machining processparameters The model underpinning the approach isdetailed by Munoz and Sheng (1995) and extended bySrinivasan and Sheng (1999a,b) based on feature-basedcase study components, exploring environmentalplanning at the micro (i.e cutting tool parameters)and macro planning (i.e setup and feature sequencing)levels (Srinivasan and Sheng 1999a,b)

Dahmus and Gutowski (2004) presented a level environmental analysis of machining processes.This work outlined an analysis together with abreakdown of energy usage for different machine tooltypes from manual machines to modern machiningcentres Further work on CAPP by He et al (2007)related to systems to support green manufacturing,with a CAPP system that takes into account optimisa-tion of energy consumption as part of the planningprocess

system-In 2009, research reported by Jin et al (2009)provides a multi-objective optimisation model forenvironmentally conscious CAPP This article outlines

a mathematical model that takes into account als, environmental data and environmental impact ofthe materials based on existing commercial databasetools to compute an environmental score for eachtooling operation This approach combined with theapproach by Xu and Li (2009) provides a basis for anew goal-oriented, multi-parameter approach to re-presenting process parameter selection at multi-levels,incorporating both micro- and macro-level decisions,and incorporating process knowledge with mathema-tical logic

materi-Based on these ideas, Newman et al (2010)outlined a framework to validate the introduction ofenergy consumption in the objectives of process

planning for CNC machining A mathematical

repre-sentation is presented and it is shown that energy

consumption can be added to multi-criteria process

planning systems as a valid objective/criterion

4.8 Commercialisation of CAPP research

In the last decade, some of the CAPP research

outcomes have also made inroads into commercial

systems Rozenfeld and Kerry (1999) presented a

solution that allows for a step-by-step introduction of

CAPP in a company In the work of Chen et al (2006),

they presented a parametric process plan that is

dependent on feature parameters The resulting system

is flexible and expandable, where a new process plan

template for a new part family can be added and

existing ones can be updated The system is also

transferable to different companies by using the

relevant process plan templates and constraints

Zhou et al (2007a,b) proposed an approach to a total

integration of CAD and CAPP based on commercial

systems with a focus on techniques such as feature

parameters and constraints extraction, feature

prece-dence tree reconstruction, technical information

pro-cessing, automatic process marking and 3D material

stock geometric model generation The implemented

system can encapsulate design and machining intents

Because of the close connection between CAD and

CAPP, and CAPP and CAM, commercial CAPP tools

are mostly embedded in the commonly used CAD/

CAM systems such as Catia1, NX1, Pro/Engineer1

and Inventor1 The CAPP functions in these systems

are in many ways less structured and non-systematic

This has hindered sweeping changes to the CAPP

functions and the technology up-take has only been

patchy and localised This said, feature technologies

have all been used for process planning Some expert

knowledge (e.g suggested optimal machining

para-meters) has been included in the decision-making

process

Most significant of all is perhaps adoption of

Internet-based technology and integrated process

planning Process plans are made available through

the Internet using XML or other languages

Integra-tion of process planning with the entire product

development process has been a common trend for

many established CAD/CAPP/CAM systems

Das-salt1’s PLM solutions ENOVIA1and CATIA1PLM

Express are integrated with the CAD/CAPP/CAM

functionalities of CATIA1, delivering a architecture

for driving collaborative innovation Bringing

so-called High Definition PLM and Teamcenter1

soft-ware to product development, Siemens NX1 7

redefines productivity with a suite of integrated

CAD, CAE, CAPP and CAM solutions Together

with Siemens advanced CNC systems, integration hasbeen extended to include manufacturing PTC1Windchill1 as a Product Data Management tool cancentrally manage many disparate IT systems (e.g.MCAD, CAPP, ERP, Visualization and Mockup)within the company

5 Recap of CAPP methods

As identified previously feature technology plays apivotal role in process planning, with the twoapproaches: feature recognition and design by features.Feature recognition is a complex and infinite domainproblem Much desired robustness of a featurerecognition system depends on how the system isstructured with respect to the problems of complexityand infinite variety of features Feature-based designallows a more direct extraction and interpretation ofdesign information from a CAD system Feature-basedapproach has been adopted by many process planningsystems, due to its ability to facilitate the representa-tion of various types of part data in a meaningful formneeded to drive automated CAPP It is though notuncommmon that feature recognition is still required

in a feature-based system

Previous surveys showed that expert systems aregenerally perceived to be useful and fitting in assistingtasks of process planning and scheduling It allows thecapturing of knowledge from experts, and is able tosimulate the problem-solving skill of a human expert in aparticular field The benefits of expert systems includepractical decisions, time gains, improved quality andmore efficient use of resources The effectiveness of anexpert system can be increased if integrated withoperations research techniques, especially with simula-tion However, an expert system’s total reliance onconsultation with human experts for knowledge acquisi-tion may lead to a rigid and biased system

The main advantages of the neural networkapproach over rule-based systems include its ability

to recognise intermediate and complex features out feeding any previous knowledge into the system,high recognition speed, ease of computation, simplicity

with-in implementation, and robustness Neural networkscan tolerate slight errors from input and mostimportant of all, they have the ability to derive rules

or knowledge through training with examples and canallow exceptions and irregularities in the knowledge/rule base A neural network enables parallel considera-tion of multiple constraints However, the neuralnetworks based inference method cannot express theinference procedure and results in an explicit manner.There is also a lack of systematic and efficient methods

to identify an appropriate training set for a specificapplication

The power of GAs comes from the fact that the

technique is much more robust than others GAs are

not guaranteed to find the global optimum solution to

a problem, though they are generally good at finding

‘acceptably good’ solutions to problems ‘acceptably

quickly’ A problem with GAs is that the genes from a

few comparatively highly fit (but not optimal)

indivi-duals may rapidly come to dominate the population,

causing it to converge on a localised solution Like

neural nets, they can learn and deal with highly

nonlinear models and noisy data They do not need

gradient information or smooth functions In both

cases, their flexibility is a drawback, since they have to

be carefully structured and coded and are also fairly

application-specific

Although tremendous effort has been made in

developing CAPP systems, the effectiveness of these

systems have not reached a satisfactory level Most

CAPP systems available today are centralised in

architecture, vertical in sequence, and off-line in data

processing It is difficult for a centralised off-line

system to make adaptive decisions in advance, without

knowing actual status of machines on the shop floor

The rapid development of agent technology could

bring CAPP research into a new era This has attracted

the attention of researchers in both CAPP and agent

technology areas Agent-based approaches offer some

unique functionality for distributed product design and

manufacturing, i.e modularity, reconfigurability,

scal-ability, upgradescal-ability, and robustness The fact that

process planning for a complex part can be broken

down into smaller planning problems, makes these

problems manageable by a number of intelligent

entities (i.e agents) working in tandem For example,

the STEP-NC environment with its clear entity

definitions and expandable architecture provides a

vast potential field for agents As full product

information is maintained in the STEP-NC

architec-ture, it can serve as an excellent knowledge base for

artificial intelligence systems Some outstanding issues

still beg answers Selection of suitable system

archi-tecture for agent organisation and an appropriate

approach for agent encapsulation remain to be tricky

tasks There is a need to design and implement effective

mechanisms and protocols for communication,

coop-eration, coordination, and negotiation It is not easy to

decide the appropriate level of the decision autonomy

of agents; nor is it easy to resolve their eventual

conflicts There is also uncertainty about the

guaran-tees (if any) that local decisions will produce and

maintain a globally acceptable behaviour Getting an

acceptable response in a timely fashion is also an open

question

Fuzzy logic and PNs have all found their

applica-tions in CAPP The fact that they have strength and

weaknesses means that combining one with someothers is likely to offer a better solution This iscertainly true with some effective combinations of thetechnologies, i.e GA combined with fuzzy logic, fuzzylogic combined with PNs to give raise to a moreeffective fuzzy PN, and integration of fuzzy logic rules,artificial neural networks, agent-based technology andexpert system

Internet-based CAPP systems are developed due tothe global competition and rapidly changing customerrequirements It provides tremendous potential forremote integration and cooperation in global manufac-turing applications It enables engineers to achieve thedynamic tool and machine selection, thus aids theexisting CAPP systems to generate realistic and eco-nomical process plans, allowing designers to efficientlyundertake manufacturability evaluation An Internet-based system also allows process planners in anyindustry to react to any unanticipated changes Tosupport the data exchange between different systemsused at different virtual enterprise companies, thecombination of the Internet and STEP provides anattractive solution However, there are still somesetbacks in the current systems, such as too large datapackage to be transferred and lack of methods that canmanage cooperative control commands more efficiently.STEP helps to solve the problem of sharinginformation between different applications and soft-ware that arise from the incompatibility of productdata representations (Kramer et al 2006) Because ofits neutral product data format, use of STEP in CAPPleads to the possibility of using standard datathroughout the entire product process chain in themanufacturing environment, hence increases the sys-tem’s capability of integrating with other systems in theentire CIM environment Publication of STEP-NCstandards has paved the way for a total integration ofCAD, CAPP, CAM and CNC by unifying the data fordesign, process planning and numerical control (New-man et al 2008)

6 A statistic analysisThis section is mainly about a statistic analysis of thenine established methods in CAPP, namely feature-based technologies, knowledge-based systems, artificialneural networks, GAs, fuzzy set theory and fuzzy logic,PNs, agent-based technology, Internet-based technol-ogy, and STEP-compliant method The publicationscited in this article and discussed in this section aremainly sought from the Elsevier’s Scopus abstract andcitation database (http://www.info.scopus.com) Thedatabase has nearly 18,000 titles from more than

5000 international publishers, including coverage of16,500 peer-reviewed journals Therefore, the statistics

presented in this section are also based on the data

from the Scopus database All the publication data was

obtained on the date of 31 December 2009 Unless they

are of any significant nature, conference publications

are not included in the analysis Only articles with a

clear focus on CAPP for machining operations are

considered in the analysis

6.1 Previous review articles

There are 10 CAPP review articles since 1984 (Table 1)

The article by Alting and Zhang (1989) published in

the International Journal of Production Research

continues to lead the charge with 143 citations as of

31 December 2009 Closely following, it is the article

(ElMaraghy et al 1993) by a group of 16 CIRP experts

and published 4 years later with 121 citations There is

then a big gap between these two articles and the rest in

terms of citations It is though worth mentioning that

the article written by Cay and Chassapis and published

in the Computers in Industry in 1997 has already been

cited 56 times There are no comprehensive CAPP

review articles since 2000

6.2 A statistic view of the publishing journals

There are close to 30 journals publishing research

articles in the areas of CAPP The journals that

published four or more CAPP-related articles in a

space of 10 years (2000–2009) are shown in Figure 3(a)

In total, these journals have produced close to 85%

of the CAPP research works published in the last 10

years As shown in the figure, both International

Journal of Production Research and InternationalJournal of Advanced Manufacturing Technology pub-lished far more extensively than any other journals did.All 10 journals included in Figure 3(a) publish varyingnumber of issue/articles each year Considering thetotal number of articles published by these journals (asshown at the end of the labels in Figure 3(b)), both theInternational Journal of Advanced ManufacturingTechnology and International Journal of ComputerIntegrated Manufacturing come out on top Thepercentiles in the figure denote the CAPP researchwork published by the journals as the proportion ofthe total number of articles published in years 2000–

2009 by the same journals These percentiles representthe so-called CAPP Subject Strength for each journalconcerned As shown in the figure, there are fourjournals having the CAPP Subject Strength higherthan 1%: International Journal of Advanced Manufac-turing Technology (2.2%), International Journal ofComputer Integrated Manufacturing (2.1%), Journal

of Intelligent Manufacturing (1.5%) and Computers inIndustry(1.1%)

Figure 4 shows the total number of CAPP researchpublications from 2000 to 2009 Though there is noclear pattern or trend to observe, it is evident that thenumber of publications has been down in the last 3years

6.3 Histograms of CAPP methodsOnce again considering the time span of 2000–2009,graphs are produced based on the number of journalpublications for each of the nine methods (Figure 5)

Table 1 CAPP previous review articles

Alting and Zhang Computer aided process planning:

The state-of-the-art survey

1989 International Journal of

Production Research

143ElMaraghy et al Evolution and future

perspectives of CAPP

1993 CIRP Annals – Manufacturing

Technology

121Cay and Chassapis An IT view on perspectives of

computer aided processplanning research

expert systems for processplanning Methods and problems

1995 International Journal of Advanced

Manufacturing Technology

44

planning: a state of art

1998 International Journal of

Advanced Manufacturing Technology

36

planning: past, present and future

1984 International Journal of

Production Research

22

computer-aided process planning

1996 International Journal of Advanced

Manufacturing Technology

13Eversheim and

planning and NC programming techniques

1991 Computer-Aided Engineering Journal 12Gupta and Ghosh A survey of expert systems in

manufacturing and process planning

Figure 4 Total number of CAPP journal publications

(2000–2009)

Figure 3 Journals publishing most CAPP-related works (a) Number of articles published on CAPP from 2000 to 2009 (b)Percentage of articles published on CAPP from 2000 to 2009

Note that some publications utilised more than one

method; these publications therefore appear in more

than one diagram

The diagrams clearly show, that over the last 3years, there is a reduction of published work usingGAs, fuzzy set theories, PN and agent-based technol-ogies It seems that PN-based methods have nevergained any strong traction over this period One canalso conclude that both Internet-based and STEP-compliant methods have recently attracted moreattention than any other methods

6.4 Technology impact factorJournal impact factor has been used as a measurereflecting the average number of citations to articlespublished in journals (http://www.isiwebofknowledge.com/) It is frequently used as a proxy for the relativeimportance of a journal within its field, with journals

Figure 5 CAPP methods histograms (2000–2009).

Figure 6 CAPP TIF for 2009

with higher impact factors deemed to be more

important than those with lower ones Likewise, the

authors defined technology impact factor (TIF) is

defined as:

‘In a given year (e.g 2009), the impact factor of a

technology is the average number of citations to those

articles that were published on the technology during

the nine preceding years (e.g 2000–2008).’

For example, the 2009 TIF for CAPP technology

A, written as TIF [2009(A)] would be calculated as

follows:

X, the number of times ‘research articles’ published

on technology A in 2000 through to 2008 were cited by

all indexed journals; Y, the total number of research

articles on technology A published in years from 2000

to 2008 (‘Research articles’ are publications on CAPP

research results, not including work loosely related to

CAPP technology reviews, notes, editorials or

Letters-to-the-Editor.)

Hence, TIF [2009(A)] ¼ X/Y

Figure 6 shows the CAPP TIF for 2009, counting

all articles published from 2000 to 2008 TIF is a

relative measure, in favour of the situation whereby

articles have a higher number of citations but the

number of articles is low This is certainly true in the

case of PN methods and maybe fuzzy set theory

methods, too TIF also favours articles that are

published in the earlier years of the time considered

This is because the longer time the article has existed

for, the more chances the article may be cited This

somehow explains why the GA and agent-based

methods have a higher TIF The argument is also

conversely backed up with both the Internet-based andSTEP-compliant methods having a lower TIF

7 Future trendsAlthough the above two sections may help predict aforeseeable future and research directions for CAPP-related research, it is never an easy task to do so in anycapacity Technologies in the domain of artificialintelligence have made a sizeable impact on CAPPresearch in the 1990s It is though not quite the case for

the first decade of the twenty-first century, with an

exception of GA-based approaches that have a higher

TIF One can safely expect CAPP research to continue

in the direction of using STEP data models and

supporting an Internet-based environment (see Figure

5) Such a trend echoes the regime in which current

manufacturing enterprises function This regime is

featured as collaborative and distributive Hence, there

is a need for the ability to share product data between

different applications at different locations, seamlessly

and collaboratively STEP will continue to play a

critical role in serving the need to share these data

between different applications STEP has been in

existence for less than 20 years and STEP-NC has

even shorter existence Another key drive for

STEP-NC to be employed is the fact that high-level

machining optimisation can be achieved at the

controller and in real time

In terms of adaptability, FBs-enabled approach

will play an active role in CAPP in response to

shop-floor uncertainty This is because that FB-embedded

algorithms can make decisions adaptively to changes at

runtime, and can be integrated with dynamic

schedul-ing and STEP-NC systems towards next-generation

adaptive and interoperable manufacturing In terms of

CAPP system implementation, the browser/server

architecture will continue to gain its popularity This

is because the Web infrastructure can facilitate

distributed collaborations at low cost for users, and

better system portability and platform independency

When using Web or other Internet-based approaches,

the secrecy of any proprietary information must be

properly maintained and managed

It is abundantly clear that there is no ‘silver bullet’

for solving CAPP problems CAPP systems that take

advantage of several technologies, e.g agent-based

approach combined with STEP data model, may offer

an attractive solution It is also not practical to expect

one CAPP system/method to be used for different

types of process planning tasks This is largely due to

the diversity exhibited by different manufacturing

processes On this note, CAPP research in other

domains has not attracted as much attention as for

metal cutting Limited work has been reported for

sheet-metal fabrication (Duflo et al 2005, Xie and Xu

2008), extrusions (Zhang et al 2006), welding

pro-cesses (Zhou et al 2007a,b), rapid prototyping

processes (Pande and Kumar 2008), and printed circuit

board (PCB) fabrication (Law and Tam 2000) The

same can be said for micro and nano-manufacturing

processes In recent years, there has been an increased

interest in developing micro-electromechanical system

(MEMS) technology for biological applications known

as bioMEMS or biochips Wang and Gong (2009)

proposed a novel automated process-planning

approach for fabrication of 3D microstructures inbioMEMS The difficulty lies in the fact that differentbioMEMSs often require totally different fabricationmethods, hence drastically different process planningtasks

As process planning is part of the entire productdevelopment process, the need for CAPP to be anintegral part of the supporting system is always there

In fact, in many cases, integrated CAPP is not adesired option but a must A CAPP system may becapable of accessing the integrated body of manufac-turing knowledge existing in the enterprise such as thatsuggested in the universal manufacturing platform byNewman and Nassehi (2007) In such an environment,the CAPP system would have access to process data,and also an integrated representation of the entirebody of the manufacturing information includingresources and product data This requires the necessity

to have a standardised way for modelling ing resources (Vichare et al 2009) Based on theseresource models, it is envisaged that the next genera-tion CAPP system has the additional ability to create acapability-adjusted process plan based on an actualresource capability instead of the nominal resourceinformation This need has been recognised by manyincluding Newman and Nassehi (2009) Process plan-ning considering issues such as energy consumptionand other environmental factors, meets the require-ments of a paradigm shift towards environmentallyconscious production The research work in this areahas commenced and is expected to be one of the keyfocused areas of CAPP research in the near future

manufactur-8 Final words

Of the two types of CAPP approaches, the variantapproach continues to be used by some manufacturingcompanies The trend though is toward a generativeapproach The feature-based approach has beenadopted by many developed process-planning systems,due to its ability to facilitate the representation ofvarious types of part data in a meaningful form needed

to drive the automated process planning There areusually a number of key steps in the development of aprocess plan It is important to note that theseactivities are inter-connected; hence, iterations arecommon activities in process planning

Expert systems are generally perceived to be useful

in process planning and scheduling as it allows thecapturing of knowledge from experts, and is able tosimulate the problem solving skill of a human expert in

a particular field Neural network, GA, fuzzy logic and

PN all belong to artificial intelligence Artificialintelligence plays an important role in CAPP, as itallows a process planning system to be adaptive and

self-learning AI techniques allow the generation of

optimum process plans/operation sequences Often

they are integrated with other technologies to produce

an intelligent system Many difficulties have been

experienced by the researchers in effectively

imple-menting an AI technique in CAPP A reasonably quiet

period of time recently (i.e 2006–2009) (Figure 5)

confirms that these difficulties still remain Publications

on the Internet-based and STEP-compliant methods

have been on the rise in recent years (Figure 5) This

trend agrees to the call for the CAPP systems to

become more collaborative and distributive to better

meet the need of a globalised manufacturing trend

When it comes to the TIF, GAs and agent-based

methods fare much better than others However, there

is a reason to believe that a similar TIF can be expected

in the next 5–10 years for both the Internet-based and

STEP-compliant methods TIF defined in this article

may be utilised elsewhere Instead of 10-year being

used for calculation, any other time span may also be

used

Out of the 10 major journals that are publishing

CAPP research works, International Journal of

Produc-tion Research and International Journal of Advanced

Manufacturing Technology published far more than

any other journals did However, considering the total

number of articles published by these journals,

International Journal of Advanced Manufacturing

Technology, International Journal of Computer

Inte-grated Manufacturing, Journal of Intelligent

Manufac-turing and Computers in Industry come out on top In

other words, they have a high CAPP Subject Strength

In authors’ humble opinion, CAPP research will

continue not only for the unsolved issues but also for

the complex and intriguing nature of the problems that

never failed to fascinate and challenge researchers

New technologies and tools will continue to be

developed, tested and hopefully adopted As the

manufacturing industry becomes more and more

globalised and mass customisation becomes a norm

for many industry sectors, CAPP systems will have to

become more adaptive, distributed, agile and

inte-grated As environmental issues become an important

factor for modern production industry to consider,

energy-conscious and energy-efficient process planning

demands more research effort For CAPP research to

be better informed, it seems that industry must also

come to the table and play a more significant role in

CAPP research, too

Acknowledgements

The authors acknowledge the contribution from Cindy Shum

and other members of the Interoperable and Intelligent

Manufacturing Systems research group at the University of

Auckland

References

Agrawal, R., et al., 2009 Multi-agent system for distributedcomputer-aided process planning problem in e-manufac-turing environment International Journal of AdvancedManufacturing Technology, 44 (5–6), 579–594

Alberti, M., Ciurana, J., and Casadesus, M., 2005 A systemfor optimising cutting parameters when planning millingoperations in high-speed machining Journal of MaterialsProcessing Technology, 168 (1), 25–35

Allen, R.D., Harding, J.A., and Newman, S.T., 2005 Theapplication of STEP-NC using agent-based processplanning International Journal of Production Research,

43 (4), 655–670

Alting, L and Zhang, H., 1989 Computer aided processplanning: the state-of-the-art survey International Jour-nal of Production Research, 27 (4), 553–585

Alvares, A.J and Ferreira, J.C.E., 2008 A system for thedesign and manufacture of feature-based parts throughthe Internet International Journal of Advanced Manu-facturing Technology, 35 (7–8), 646–664

Alvares, A.J., Ferreira, J.C.E., and Lorenzo, R.M., 2008 Anintegrated web-based CAD/CAPP/CAM system for theremote design and manufacture of feature-based cylind-rical parts Journal of Intelligent Manufacturing, 19 (6),643–659

Amaitik, S.M and Kilic, S.E., 2005 STEP-based featuremodeller for computer-aided process planning Interna-tional Journal of Production Research, 43 (15), 3087–3101

Amaitik, S.M and Kilic, S.E., 2007 An intelligent processplanning system for prismatic parts using STEP features.International Journal of Advanced Manufacturing Tech-nology, 31 (9–10), 978–993

Anwer, N and Chep, A., 1999 IPPA: a knowledge assistedprocess planning system In: IEEE Symposium onEmerging Technologies and Factory Automation, 7thIEEE International Conference on Emerging Technologiesand Factory Automation (ETFA099),ETFA, Barcelona,Spain, IEEE

Arezoo, B., Ridgway, K., and Al-Ahmari, A.M.A., 2000.Selection of cutting tools and conditions of machiningoperations using an expert system Computers in Industry,

42 (1), 43–58

Babic, B., Nesic, N., and Miljkovic, Z., 2008 A review ofautomated feature recognition with rule-based patternrecognition Computers in Industry, 59 (4), 321–337.Balic, J and Korosec, M., 2002 Intelligent tool pathgeneration for milling of free surfaces using neuralnetworks International Journal of Machine Tools andManufacture, 42 (10), 1171–1179

Beg, J and Shunmugam, M.S., 2003 Application of fuzzylogic in the selection of part orientation and probeorientation sequencing for prismatic parts InternationalJournal of Production Research, 41 (12), 2799–2815.Ben Yahia, N., et al., 2002 Manufacturing process planningapplication using artificial neural networks In: Proceed-ings of the IEEE International Conference on Systems,Man and Cybernetics, Vol 5649–654

Bhaskara Reddy, S.V., Shunmugam, M.S., and Narendran,T.T., 1999 Operation sequencing in CAPP using geneticalgorithms International Journal of Production Research,

37 (5), 1063–1074

Bo, Z.W., Hua, L.Z., and Yu, Z.G., 2006 Optimization ofprocess route by genetic algorithms Robotics andComputer-Integrated Manufacturing, 22 (2), 180–188

Brecher, C., Vitr, M., and Wolf, J., 2006 Closed-loop CAPP/

CAM/CNC process chain based on STEP and STEP-NC

inspection tasks International Journal of Computer

Integrated Manufacturing, 19 (6), 570–580

Cai, N., Wang, L., and Feng, H.-Y., 2009 GA based

adaptive setup planning toward process planning and

scheduling integration International Journal of

Produc-tion Research, 47 (10), 2745–2766

Canales, J.C., Li, X., and Yu, W., 2006 Fuzzy knowledge

learning via adaptive Fuzzy Petri Net with Triangular

function model In: Proceedings of the World Congress on

Intelligent Control and Automation (WCICA),Dalian

Carpenter, I.D and Maropoulos, P.G., 2000 Flexible tool

selection decision support system for milling operations

Journal of Materials Processing Technology, 107 (1–3),

143–152

Cay, F and Chassapis, C., 1997 An IT view on perspectives

of computer aided process planning research Computers

in Industry, 34 (3), 307–337

Chan, F.T.S., Zhang, J., and Li, P., 2001 Modelling of

integrated, distributed and cooperative process planning

system using an agent-based approach Proceedings of

Instrumentation and Mechanical Engineering, Part B – Journal

of Engineering Manufacturing, 215 (B10), 1437–1451

Chang, P.T and Chang, C.H., 2000 An integrated artificial

intelligent computer-aided process planning system

International Journal of Computer Integrated

Manufac-turing, 13 (6), 483–497

Chen, Y., et al., 2006 Parametric process planning based on

feature parameters of parts International Journal of

Advanced Manufacturing Technology, 28 (7–8), 727–736

Chiung, M., Yin-Zhen, L., and Gen, M., 1998 Evolutionary

algorithm for flexible process sequencing with multiple

objectives Proceedings of the IEEE Conference on

Evolu-tionary Computation, ICEC 4–9 May Anchorage, AK,

USA Washington D.C.: IEEE Press, 27–32

Chung, C and Peng, Q., 2004 The selection of tools and

machines on web-based manufacturing environments

International Journal of Machine Tools and Manufacture,

44 (2–3), 317–326

Chung, D.H and Suh, S.H., 2008 ISO 14649-based nonlinear

process planning implementation for complex machining

CAD Computer Aided Design, 40 (5), 521–536

Ciurana, J., et al., 2008 A model for integrating process

planning and production planning and control in

machining processes Robotics and Computer-Integrated

Manufacturing, 24 (4), 532–544

Culler, D.E and Burd, W., 2007 A framework for extending

computer aided process planning to include business

activities and computer aided design and manufacturing

(CAD/CAM) data retrieval Robotics and

Computer-Integrated Manufacturing, 23 (3), 339–350

Dahmus, J.B and Gutowski, T.G., 2004 American Society

of Mechanical Engineers Manufacturing Engineering

Division, MED, Anaheim, California, USA

IM-ECE2004-62600, 643–652

Dashora, Y., Tiwari, M.K., and Karunakaran, K.P., 2008 A

psycho-clonal-algorithm-based approach to the solve

operation sequencing problem in a CAPP environment

International Journal of Computer Integrated

Manufac-turing, 21 (5), 510–525

Deb, S., Ghosh, K., and Paul, S., 2006 A neural network

based methodology for machining operations selection in

computer-aided process planning for rotationally

sym-metrical parts Journal of Intelligent Manufacturing, 17

(5), 557–569

Denkena, B., et al., 2007 Knowledge management in processplanning CIRP Annals – Manufacturing Technology, 56(1), 175–180

De´pince´, P., Amara, H., and Hascoe¨t, J.-Y., 2001 agent environment for process planning elaboration.IEEE Symposium on Emerging Technologies andFactory Automation, ETFA, Antibes-Juan les pins, 2,783–786

Multi-Dereli, T and Filiz, I.H., 1999 Optimisation of processplanning functions by genetic algorithms Computers andIndustrial Engineering, 36 (2), 281–308

Devireddy, C.R and Ghosh, K., 1999 Feature-basedmodelling and neural networks-based CAPP for inte-grated manufacturing International Journal of ComputerIntegrated Manufacturing, 12 (1), 61–74

Devireddy, C.R., Eid, T., and Ghosh, K., 2002 aided process planning for rotational components usingartificial neural networks International Journal of AgileManufacturing, 5 (1), 27–49

Computer-Ding, L., et al., 2005 Global optimization of a feature-basedprocess sequence using GA and ANN techniques Interna-tional Journal of Production Research, 43 (15), 3247–3272.Duflou, J.R., et al., 2005 Automated tool selection forcomputer-aided process planning in sheet metal bending.CIRP Annals – Manufacturing Technology, 54 (1), 451–454

ElMaraghy, H.A., et al., 1993 Evolution and futureperspectives of CAPP CIRP Annals – ManufacturingTechnology, 42 (2), 739–751

Eversheim, W and Schneewind, J., 1993 Computer-aidedprocess planning – state of the art and future develop-ment Robotics and Computer-Integrated Manufacturing,

10 (1–2), 65–70

Eversheim, W and Schulz, J., 1985 CIRP technical reports:survey of computer aided process planning systems.CIRP Annals, 34 (2), 607–613

Fernandes, K.J and Raja, V.H., 2000 Incorporated toolselection system using object technology InternationalJournal of Machine Tools and Manufacture, 40 (11),1547–1555

Ferreira, J.C.E and Wysk, R.A., 2001 An investigation of theinfluence of alternative process plans on equipment control.Journal of Manufacturing Systems, 19 (6), 393–406.Fichtner, D., et al., 2006 Use of agents and neural networksfor acquisition and preparation of distributed NC infor-mation to support NC planning International Journal ofComputer Integrated Manufacturing, 19 (6), 581–592.Frank, M.C., Wysk, R.A., and Joshi, S.B., 2006 Determin-ing setup orientations from the visibility of slice geometryfor rapid computer numerically controlled machining.Journal of Manufacturing Science and Engineering,Transactions of the ASME, 128 (1), 228–238

Gologlu, C., 2004 A constraint-based operation sequencingfor a knowledge-based process planning Journal ofIntelligent Manufacturing, 15 (4), 463–470

Gonzalez, F and Rosado, P., 2003 General and flexiblemethodology and architecture for CAPP: GF-CAPPsystem International Journal of Production Research, 41(12), 2643–2662

Gonzalez, F and Rosado, P., 2004 General informationmodel for representing machining features in CAPPsystems International Journal of Production Research, 42(9), 1815–1842

Gouda, S and Taraman, K., 1989 CAPP: AAST, presentand future Technical Paper Society of ManufacturingEngineers MS

Grabowik, C and Knosala, R., 2003 The method of

knowledge representation for a CAPP system Journal

of Materials Processing Technology, 133 (1–2), 90–98

Gupta, T and Ghosh, B.K., 1989 A survey of expert systems

in manufacturing and process planning Computers in

Industry, 11, 195–204

Halevi, G and Wang, K., 2007 Knowledge based

manu-facturing system (KBMS) Journal of Intelligent

Manu-facturing, 18 (4), 467–474

Ham, I and Lu, S.C.-Y., 1988 Computer-aided process

planning: the present and the future CIRP Annals

-Manufacturing Technology, 37 (2), 591–601

Hamou, S., et al., 2006 A stochastic concept for the

optimization of manufacturing tolerances in computer

aided process plan simulation International Journal of

Computer Integrated Manufacturing, 19 (7), 663–675

Han, J and Han, I., 1999 Manufacturable feature recognition

and its integration with process planning Proceedings of

the Symposium on Solid Modeling and Applications, 9–11

June, Ann Arbor, Michigan, USA, 108–118

He, Y., et al., 2007 Process planning support system for

green manufacturing and its application Frontiers of

Mechanical Engineering in China, 2 (1), 104–109

Hebbal, S.S and Mehta, N.K., 2008 Setup planning for

machining the features of prismatic parts International

Journal of Production Research, 46 (12), 3241–3257

Hopfield, J.J and Tank, D.W., 1985 Neural computation of

decisions in optimization Problems Biological

Cyber-netics, 52, 141–152

Hou, M and Faddis, T.N., 2006 Automatic tool path

generation of a feature-basedCAD/CAPP/CAM

inte-grated system International Journal of Computer

Inte-grated Manufacturing, 19 (4), 350–358

Hu, C.J., Li, Z.Z., Zheng, L., Li, N., and Wen, P.H., 2008

An XML-based implementation of manufacturing route

sheet documents for context-sensitive and web-based

process planning International Journal of Computer

Integrated Manufacturing, 21, 647–656

Huang, L., et al., 2008 Rule-fused technology for typical

processes representation and automatic process decision

In: IEEE/ASME International Conference on Advanced

Intelligent Mechatronics, AIM, 2–5 July, Xian, China,

1367–1372

Huang, S.H and Xu, N., 2003 Automatic set-up planning for

metal cutting: an integrated methodology International

Journal of Production Research, 41 (18), 4339–4356

Huang, Z and Yip-Hoi, D., 2002 High-level feature

recognition using feature relationship graphs CAD

Computer Aided Design, 34 (8), 561–582

Hwang, J.L and Henderson, M.R., 1992 Applying the

perception to three-dimensional feature recognition

Journal of Design and Manufacturing, 2 (4), 187–198

Hwang, J.S and Miller, W.A., 1997 Using mixed-type

reasoning in computer-aided process planning for feature

interactions Journal of Intelligent Manufacturing, 8 (4),

297–306

ISO 10303-1, 1994 Industrial automation systems and

integra-tion – product data representaintegra-tion and exchange – Part 1:

overview and fundamental principles Geneva, Switzerland:

International Organisation for Standardisation (ISO)

ISO 10303-203, 1994 Industrial automation systems and

integration – product data representation and exchange –

Part 203: application protocol: configuration controlled

3D design of mechanical parts and assemblies (modular

version) Geneva, Switzerland: International

Organisa-tion for StandardisaOrganisa-tion (ISO)

ISO 10303-224, 1999 Industrial automation systems andintegration – Product data representation and exchange –Part 224: application protocol: mechanical productdefinition for process planning using machining features.Geneva, Switzerland: International Organisation forStandardisation (ISO)

ISO 10303-214, 2001 Industrial automation systems andintegration – product data representation and ex-change – Part 214: application protocol: core datafor automotive mechanical design processes Geneva,Switzerland: International Organisation for Standardi-sation (ISO)

ISO 14649-1, 2004 Industrial automation systems andintegration – physical device control – data model forcomputerized numerical controllers – Part 1: overviewand fundamental principles Geneva, Switzerland: Inter-national Organisation for Standardisation (ISO).ISO 14649-10, 2004 Industrial automation systems andintegration – physical device control – data model forcomputerized numerical controllers – Part 10: generalprocess data Geneva, Switzerland: International Orga-nisation for Standardisation (ISO)

ISO 14649-11, 2004 Industrial automation systems andintegration – physical device control – data model forcomputerized numerical controllers – Part 11: processdata for milling Geneva, Switzerland: InternationalOrganisation for Standardisation (ISO)

ISO 10303-238, 2007 Industrial automation systems andintegration – product data representation and exchange –Part 238: application protocol: application interpretedmodel for computerized numerical controllers Geneva,Switzerland: International Organisation for Standardisa-tion (ISO)

Jang, P.Y., Son, Y.J., and Cho, H., 2003 Elaboration andvalidation of AND/OR graph-based non-linear processplans for shop floor control International Journal ofProduction Research, 41 (13), 3019–3043

Jia, X.-L., et al., 2003 The research on representation andprocessing of process knowledge based on object-oriented modelling International Conference on MachineLearning and Cybernetics, 2, 657–660

Jiang, B., et al., 1999 An automatic process planning systemfor the quick generation of manufacturing process plansdirectly from CAD drawings Journal of MaterialsProcessing Technology, 87 (1–3), 97–106

Jin, K., et al., 2009 A multiple objective optimization modelfor environmental benign process planning In: Proceed-ings 2009 IEEE 16th International Conference on In-dustrial Engineering and Engineering Management 21–23October, New York: IEEE Beijing, China, 869–873.Joneja, A and Chang, T.-C., 1999 Setup and fixtureplanning in automated process planning systems IIETransactions (Institute of Industrial Engineers), 31 (7),653–665

Joo, J., et al., 2001 Dynamic planning model for determiningcutting parameters using neural networks in feature-based process planning Journal of Intelligent Manufac-turing, 12 (1), 13–29

Kamrani, A.K., Sferro, P., and Handelman, J., 1995 Criticalissues in design and evaluation of computer aided processplaning systems Computers and Industrial Engineering,

29 (1–4), 619–623

Kang, M., Han, J., and Moon, J.G., 2003 An approach forinterlinking design and process planning Journal ofMaterials Processing Technology, 139 (1–3 SPEC), 589–595

Kasirolvalad, Z., Motlagh, M.R.J., and Shadmani, M.A., 2004.

An intelligent modular modelling approach for quality

control of CNC machines product using adaptive fuzzy

Petri nets In: 2004 8th International Conference on Control,

Automation, Robotics and Vision (ICARCV), Kunming

Khoshnevis, B., Sormaz, D.N., and Park, J.Y., 1999 An

integrated process planning system using feature

reason-ing and space search-based optimization IIE

Transac-tions (Institute of Industrial Engineers), 31 (7), 597–616

Kiritsis, D., 1995 Review of knowledge-based expert systems

for process planning Methods and problems

Interna-tional Journal of Advanced Manufacturing Technology, 10

(4), 240–262

Kiritsis, D., Neuendor, K.P., and Xirouchakis, P., 1999 Petri

net techniques for process planning cost estimation

Advances in Engineering Software, 30 (6), 375–387

Korosec, M., et al., 2005 Neural network based

manufactur-ability evaluation of free form machining International

Journal of Machine Tools and Manufacture, 45 (1), 13–20

Kramer, T.R., et al., 2006 Run-time interpretation of

STEP-NC: implementation and performance International

Journal of Computer Integrated Manufacturing, 19 (6),

495–507

Krishna, A.G and Rao, K.M., 2006 Simultaneous optimal

selection of design and manufacturing tolerances with

different stack-up conditions using scatter search

Inter-national Journal of Advanced Manufacturing Technology,

30 (3–4), 328–333

Kumar, M and Rajotia, S., 2005 Development of a

generative CAPP system for axisymmetric components

for a job shop environment International Journal of

Advanced Manufacturing Technology, 27 (1–2), 136–144

Lau, H.C.W., et al., 2005 Development of a

computer-integrated system to support CAD to CAPP

Interna-tional Journal of Advanced Manufacturing Technology, 26

(9–10), 1032–1042

Law, H.W and Tam, H.Y., 2000 Object-oriented analysis

and design of computer aided process planning systems

International Journal of Computer Integrated

Manufac-turing, 13 (1), 40–49

Lee, D.H., Kiritsis, D., and Xirouchakis, P., 2004 Iterative

approach to operation selection and sequencing in

process planning International Journal of Production

Research, 42 (22), 4745–4766

Lee, H.C., Jhee, W.C., and Park, H.S., 2007 Generative

CAPP through projective feature recognition Computers

and Industrial Engineering, 53 (2), 241–246

Leung, H.C., 1996 Annotated bibliography on

computer-aided process planning International Journal of Advanced

Manufacturing Technology, 12 (5), 309–329

Li, L., et al., 2005 Application of genetic algorithm to

computer-aided process planning in distributed

manu-facturing environments Robotics and

Computer-Inte-grated Manufacturing, 21 (6), 568–578

Li, W., et al., 2000a Optimization of machining datum

selection and machining tolerance allocation with genetic

algorithms International Journal of Production Research,

38 (6), 1407–1424

Li, W.D., Ong, S.K., and Nee, A.Y.C., 2002 Hybrid genetic

algorithm and simulated annealing approach for the

optimization of process plans for prismatic parts

Interna-tional Journal of Production Research, 40 (8), 1899–1922

Li, X., et al., 2000b Dynamic knowledge inference and

learning under adaptive fuzzy Petri net framework IEEE

Transactions on Systems, Man and Cybernetics Part C:

Applications and Reviews, 30 (4), 442–450

Liao, S.H., 2005 Expert system methodologies and tions – a decade review from 1995 to 2004 ExpertSystems with Applications, 28 (1), 93–103

applica-Liu, C., et al., 2004 A component-based computer aidedprocess planning system for CSCW Proceedings The 8thInternational Conference on Computer Supported Coop-erative Work in Design, 1 (1), 169–173

Liu, S., Zhang, Z., and Tian, X., 2007 A typical processroute discovery method based on clustering analysis.International Journal of Advanced Manufacturing Tech-nology, 35 (1–2), 186–194

Liu, W and Peng, G., 2005 An internet-enabled setupplanning system In: Proceedings – 3rd InternationalConference on Information Technology and Applications,ICITA 2005, Sydney

Liu, Z and Wang, L., 2007 Sequencing of interactingprismatic machining features for process planning.Computers in Industry, 58 (4), 295–303

Markus, A., Vancza, J., and Horvath, M., 1997 Processplanning by retrieval and adaptation Computers inIndustry, 33 (1), 47–60

Marri, H.B., Gunasekaran, A., and Grieve, R.J., 1998.Computer-aided process planning: a state of art Inter-national Journal of Advanced Manufacturing Technology,

14 (4), 261–268

Meseguer, A and Gonzalez, F., 2008 A methodology forcutting-tool management through the integration ofCAPP and scheduling International Journal of Produc-tion Research, 46 (6), 1685–1706

Metaxiotis, K.S., Askounis, D., and Psarras, J., 2002a.Expert systems in production planning and scheduling: astate-of-the-art survey Journal of Intelligent Manufactur-ing, 13 (4), 253–260

Min, L., Li, B., and Zhang, S., 2004 Modeling integratedCAPP/PPS systems Computers and Industrial Engineer-ing, 46 (2), 275–283

Ming, X.G and Mak, K.L., 2000a A hybrid Hopfieldnetwork-genetic algorithm approach to optimal processplan selection International Journal of Production Re-search, 38 (8), 1823–1839

Ming, X.G and Mak, K.L., 2000b Intelligent setupplanning in manufacturing by neural networks basedapproach Journal of Intelligent Manufacturing, 11 (3),311–331

Ming, X.G and Mak, K.L., 2001 Intelligent approaches totolerance allocation and manufacturing operations selec-tion in process planning Journal of Materials ProcessingTechnology, 117 (1–2), 75–83

Ming, X.G., Mak, K.L., and Yan, J.Q., 1999 A hybridintelligent inference model for computer aided processplanning Integrated Manufacturing Systems, 10 (6), 343–353

Munoz, A.A and Sheng, P., 1995 An analytical approachfor determining the environmental impact of machiningprocesses Journal of Materials Processing Technology, 53(3–4), 736–758

Nassehi, A., Liu, R., and Newman, S.T., 2007 A newsoftware platform to support feature-based processplanning for interoperable STEP-NC manufacture.International Journal of Computer Integrated Manufac-turing, 20 (7), 669–683

Nassehi, A., Newman, S.T., and Allen, R.D., 2006 Theapplication of multi-agent systems for STEP-NC com-puter aided process planning of prismatic components.International Journal of Machine Tools and Manufacture,

46 (5), 559–574

Newman, S.T., et al., 2008 Strategic advantages of

interoperability for global manufacturing using CNC

technology Robotics and Computer-Integrated

Manufac-turing, 24 (6), 699–708

Newman, S.T and Nassehi, A., 2007 Universal

manufactur-ing platform for CNC manufacture Annals of CIRP, 56

(1), 459–462

Newman, S.T and Nassehi, A., 2009 Machine tool

capability profile for intelligent process planning CIRP

Annals – Manufacturing Technology, 58 (1), 421–424

Newman, S.T., Nassehi, A., and Imani-Asrai, R., 2010 Energy

efficient process planning for CNC machining In:

Pro-ceedings of 1st International Colloquium of the Cluster of

Excellence Energy-Efficient Product and Process

Innova-tion in ProducInnova-tion Engineering (eniPROD), Pubs R

Neugebauer, Technical University of Chemnitz, Germany,

57–80

Niebel, B.W., 1965 Mechanized process selection for

planning new designs ASME 33rd Annual Meeting

collected papers V65 Book 4, Paper No 737

Ong, S.K., Li, W.D., and Nee, A.Y.C., 2003 STEP-based

integration of feature recognition and design-by-feature

for manufacturing applications in a concurrent

engineer-ing environment International Journal of Computer

Applications in Technology, 18 (1–4), 78–92

Onwubolu, G.C., 1999 Manufacturing features recognition

using backpropagation neural networks Journal of

Intelligent Manufacturing, 10 (3), 289–299

Pan, S., Jiang, S., and Xu, Y., 2002 A cost estimation model

of product based on CAD/CAPP integration

Proceed-ings of the World Congress on Intelligent Control and

Automation (WCICA), 4, 2650–2654

Pande, S.S and Kumar, S., 2008 A generative process

planning system for parts produced by rapid

prototyp-ing International Journal of Production Research, 46 (22),

6431–6460

Park, M.W., et al., 2000 Incremental supervised learning of

cutting conditions using the fuzzy ARTMAP neural

network CIRP Annals - Manufacturing Technology, 49

(1), 375–378

Park, B.T., Park, M.W., and Kim, S.-K., 2001 Generation

and evolutionary learning of cutting conditions for

milling operations International Journal of Advanced

Manufacturing Technology, 17 (12), 870–880

Park, S.C., 2003 Knowledge capturing methodology in

process planning CAD Computer Aided Design, 35 (12),

1109–1117

Patil, L and Pande, S.S., 2002 An intelligent feature-based

process planning system for prismatic parts International

Journal of Production Research, 40, 4431–4447

Peng, G., Liu, W., and Zhang, X., 2005 An Internet-based

system for setup planning in machining operations In:

Proceedings of the IEEE International Conference on

Engineering of Complex Computer Systems, ICECCS,

Shanghai, 245–251

Peng, Q., Hall, F.R., and Lister, P.M., 2000 Application and

evaluation of VR-based CAPP system Journal of

Materials Processing Technology, 107 (1–3), 153–159

Pham, D.T and Gologlu, C., 2001 A computer aided process

planning system for concurrent engineering Proceedings

of the Institution of Mechanical Engineers, Part B: Journal

of Engineering Manufacture, 215 (8), 1117–1131

Prabhakar, S and Henderson, M.R., 1992 Automatic

form-feature recognition using neural network-based

techni-ques on B-rep of solid models Computer Aided Design,

24 (7), 381–393

Qiao, L., Wang, X.Y., and Wang, S.-C., 2000 A GA-basedapproach to machining operation sequencing for pris-matic parts International Journal of Production Research,

38 (14), 3283–3303

Qin, B., Jiang, S., and Wang, N., 2005 based combinatorial optimization method for alternativeprocess plans Zhongguo Jixie Gongcheng/China Mechan-ical Engineering, 16 (12), 1076–1079

Genetic-algorithms-Qiu, Z.M., et al., 2001 Multi-user NC machining simulationover the WWW International Journal of AdvanceManufacturing Technology, 18, 1–6

Rameshbabu, V and Shunmugam, M.S., 2009 Hybridfeature recognition method for setup planning fromSTEP AP–203 Robotics and Computer Integrated Man-ufacturing, 25 (2), 393–408

Ranjan, R., et al., 2005 Automatic recognition of machiningfeatures from a solid model using the 2D feature pattern.International Journal of Advanced Manufacturing Tech-nology, 26 (7–8), 861–869

Rocha, J., Ramos, C., and Vale, Z., 1999 Process planningusing a genetic algorithm approach Proceedings of theIEEE International Symposium on Assembly and TaskPlanning, 21–24 July, Porto, Portugal, 338–343

Roucoules, L., Salomons, O., and Paris, H., 2003 Processplanning as an integration of knowledge in the detaileddesign phase International Journal of Computer Inte-grated Manufacturing, 16 (1), 25–37

Rozenfeld, H and Kerry, H.T., Jr, 1999 Automated processplanning for parametric parts International Journal ofProduction Research, 37 (17), 3981–3993

Sadaiah, M., et al., 2002 A generative computer-aidedprocess planning system for prismatic components.International Journal of Advanced Manufacturing Tech-nology, 20 (10), 709–719

Salehi, M and Tavakkoli-Moghaddam, R., 2009 Application

of genetic algorithm to computer-aided process planning

in preliminary and detailed planning Engineering tions of Artificial Intelligence, 22 (8), 1179–1187

Applica-Saygin, C and Kilic, S.E., 1999 Integrating flexible processplans with scheduling in flexible manufacturing systems.International Journal of Advanced Manufacturing Tech-nology, 15 (4), 268–280

Shah, J.J., 1991 Assessment of features technology ter Aided Design, 23 (5), 331–334

Compu-Shah, J., Sreevalsan, P., and Mathew, A., 1991 Survey of CAD/feature-based process planning and NC programming tech-niques Computer-Aided Engineering Journal, 8 (1), 25–33.Shen, W and Norrie, D.H., 1999 Agent-based systems forintelligent manufacturing: a state-of-the-art survey.Knowledge and Information Systems, 1 (2), 129–156.Shen, W., Wang, L., and Hao, Q., 2006a Agent-baseddistributed manufacturing process planning and schedul-ing: a state-of-the-art survey Systems, Man and Cyber-netics, Part C: Applications and Reviews, IEEETransactions, 36 (4), 563–577

Shen, W., et al., 2006b Applications of agent-based systems

in intelligent manufacturing: an updated review vanced Engineering Informatics, 20 (4), 415–431.Sheng, P., Srinivasan, M., and Kobayashi, S., 1995 Multi-objective process planning in environmentally consciousmanufacturing: a feature-based approach CIRP Annals,

Ad-44 (1), 433–437

Shin, J., et al., 2001 An investigation of the influence ofalternative process plans in a dynamic shop floor environ-ment Proceedings of the IEEE International Conference onSystems, Man and Cybernetics, 2, 1304–1309

Shunmugam, M.S., Bhaskara Reddy, S.V., and Narendran,

T.T., 2000 Selection of optimal conditions in multi-pass

face-milling using a genetic algorithm International Journal

of Machine Tools and Manufacture, 40 (3), 401–414

Shunmugam, M.S., Mahesh, P., and Bhaskara Reddy, S.V.,

2002 A method of preliminary planning for rotational

components with C-axis features using genetic algorithm

Computers in Industry, 48 (3), 199–217

Singh, P.K., Jain, P.K., and Jain, S.C., 2003 Simultaneous

optimal selection of design and manufacturing tolerances

with different stack-up conditions using genetic

algo-rithms International Journal of Production Research, 41

(11), 2411–2429

Sormaz, D.N and Khoshnevis, B., 1995 Knowledge

representation for automated process planning In:

Proceedings of the IEEE International Symposium on

Assembly and Task Planning,Pittsburgh, PA

Sormaz, D.N and Khoshnevis, B., 2003 Generation of

alternative process plans in integrated manufacturing

systems Journal of Intelligent Manufacturing, 14 (6),

509–526

Sridharan, N and Shah, J.J., 2004 Recognition of multi axis

milling features – Part I: topological and geometric

characteristics Journal of Computing and Information

Science in Engineering, 4 (3), 242–250

Srinivasan, M and Sheng, P., 1999a Feature-based process

planning for environmentally conscious machining – Part

1: microplanning Robotics and Computer Integrated

Manufacturing, 15 (3), 257–270

Srinivasan, M and Sheng, P., 1999b Feature-based process

planning for environmentally conscious machining – Part

2: macroplanning Robotics and Computer Integrated

Manufacturing, 15 (3), 271–281

Steudel, H.J., 1984 Computer-aided process planning: past,

present and future International Journal of Production

Research, 22 (2), 253–266

Stroud, I and Xirouchakis, P., 2006 Strategy features for

communicating aesthetic shapes for manufacturing

International Journal of Computer Integrated

Manufac-turing, 19 (6), 639–649

Sun, J., Zhang, Y.F., and Nee, A.Y.C., 2001 A distributed

multi-agent environment for product design and

manu-facturing planning International Journal of Production

Research, 39 (4), 625–645

Sunil, V.B and Pande, S.S., 2008 Automatic recognition of

machining features using artificial neural networks The

International Journal of Advanced Manufacturing

Tech-nology, 41 (9–10), 932–947

Tian, X.-T., et al., 2007 Non-linear process planning and

optimization technique in STEP-NC-based CAPP

Ji-suanji Jicheng Zhizao Xitong/Computer Integrated

Man-ufacturing Systems, CIMS, 13 (2), 228–233

Tu, Y., Chu, X., and Yang, W., 2000 Computer-aided

process planning in virtual one-of-a-kind production

Computers in Industry, 41 (1), 99–110

Valin’o, G., et al., 2007 Methodology for set-up planning

automation of turned parts International Journal of

Production Research, 45 (17), 3917–3947

Vichare, P., et al., 2009 A unified manufacturing resource

model for representation of CNC machining systems

elements Robotics and Computer-Integrated

Manufactur-ing, 25 (6), 999–1007

Vidal, A., et al., 2005 A decision support system for

optimising the selection of parameters when planning

milling operations International Journal of Machine

Tools and Manufacture, 45 (2), 201–210

Wang, C.C.L., Chen, S.F., and Yuen, M.M.F., 2001 Fuzzypart family formation based on grey relational analysis.International Journal of Advanced Manufacturing Tech-nology, 18 (2), 128–132

Wang, H., Xu, X., and Tedford, J.D., 2007 An AdaptableCNC System based on STEP-NC and function blocks.International Journal of Production Research, 45 (17),3809–3829

Wang, L., 1997 A course in fuzzy systems and control.Englewood Cliffs: Prentice-Hall

Wang, L., Feng, H.-Y., and Cai, N., 2003 Architecturedesign for distributed process planning Journal ofManufacturing Systems, 22 (2), 99–115

Wang, L., et al., 2006 Enriched machining feature basedreasoning for generic machining process sequencing.International Journal of Production Research, 44 (8),1479–1501

Wang, L., et al., 2009 Function block design for adaptiveexecution control of job shop machining operations.International Journal of Production Research, 47 (12),3413–3434

Wang, L., Keshavarzmanesh, S., and Feng, H.-Y., 2008.Design of adaptive function blocks for dynamic assemblyplanning and control Journal of Manufacturing Systems,

27 (1), 45–51

Wang, L and Shen, W., 2003 DPP: an agent-basedapproach for distributed process planning Journal ofIntelligent Manufacturing, 14 (5), 429–439

Wang, Q.-H and Gong, H.-Q., 2009 Computer-aidedprocess planning for fabrication of three-dimensionalmicrostructures for bioMEMS applications InternationalJournal of Production Research, 47 (21), 6051–6067.Welband, M., 1983 A review of knowledge acquisitiontechniques for expert systems British Telecommunica-tions Research Laboratories Technical Report, Ipswich.Wong, T.N., Chan, L.C.F., and Lau, H.C.W., 2003.Machining process sequencing with fuzzy expert systemand genetic algorithms Engineering with Computers, 19(2–3), 191–202

Woo, Y., et al., 2005 A hybrid feature recognizer formachining process planning systems CIRP Annals –Manufacturing Technology, 54 (1), 397–400

Wu, R.R., et al., 2002 Optimal operation planning usingfuzzy Petri nets with resource constraints InternationalJournal of Computer Integrated Manufacturing, 15 (1),28–36

Xie, S.Q and Xu, X., 2008 STEP-compliant processplanning system for compound sheet metal machining.International Journal of Production Research, 46 (1), 25–50

Xu, H and Li, D., 2009 Modeling of process parameterselection with mathematical logic for process planning.Robotics and Computer Integrated Manufacturing, 25 (3),529–535

Xu, H.M and Li, D.B., 2008a A meta-modeling paradigm

of the manufacturing resources using mathematical logicfor process planning International Journal of AdvancedManufacturing Technology, 36 (9–10), 1022–1031

Xu, H.M and Li, D.B., 2008b A clustering-based modelingscheme of the manufacturing resources for processplanning International Journal of Advanced Manufactur-ing Technology, 38 (1–2), 154–162

Xu, H.M., Mi, L.C., and Li, D.B., 2008 A decision logicschema of process planning using backward chainingreasoning International Journal of Advanced Manufac-turing Technology, 38 (11–12), 1181–1191

Xu, H.M., Yuan, M.H., and Li, D.B., 2009 A novel process

planning schema based on process knowledge

customiza-tion International Journal of Advanced Manufacturing

Technology, 44 (1–2), 161–172

Xu, X and Hinduja, S., 1997 Determination of finishing

features in 2½D components Proceedings of the

Institu-tion of Mechanical Engineers, Part B: Journal of

Engineering Manufacture, 211 (2), 125–142

Xu, X and Hinduja, S., 1998 Recognition of rough

machining features in 2½D components CAD Computer

Aided Design, 30 (7), 503–516

Xu, X.W and He, Q., 2004 Striving for a total integration of

CAD, CAPP, CAM and CNC Robotics and

Computer-Integrated Manufacturing, 20 (2), 101–109

Xu, X.W., et al., 2005a STEP-compliant NC research: the

search for intelligent CAD/CAPP/CAM/CNC

integra-tion International Journal of Production Research, 43

(17), 3703–3743

Xu, Z., et al., 2005b Study of networked manufacturing

oriented cooperative CAPP system In: Proceedings of the

9th International Conference on Computer Supported

Cooperative Work in Design,Coventry

Yao, S., et al., 2007 Computer aided manufacturing

planning for mass customization – Part 2: automated

setup planning International Journal of Advanced

Man-ufacturing Technology, 32 (1–2), 205–217

Yifei, T., et al., 2008 A feature-extraction-based

process-planning system International Journal of Advanced

Manufacturing Technology, 38 (11–12), 1192–1200

Yongtao, H and Jingying, M., 2006 A knowledge-based

auto-reasoning methodology in hole-machining process

planning Computers in Industry, 57 (4), 297–304

You, C.F and Lin, C.H., 2005 Java-based computer-aided

process planning International Journal of Advanced

Manufacturing Technology, 26 (9–10), 1063–1070

You, C.F., Sheen, B.T., and Lin, T.K., 2007 Selecting

optimal tools for arbitrarily shaped pockets International

Journal of Advanced Manufacturing Technology, 32 (1–2),

132–138

Yue, Y., et al., 2002 Study of neural network techniques for

computer integrated manufacturing Engineering

Compu-tations (Swansea, Wales), 19 (1–2), 136–157

Yuen, C.F., Wong, S.Y., and Venuvinod, P.K., 2003

Development of a generic computer-aided process

plan-ning support system Journal of Materials Processing

Technology, 139 (1–3 SPEC), 394–401

Yusof, Y., et al., 2009 Interoperable CNC system for turning

operations Proceedings of World Academy of Science,

Engineering and Technology, 34, 941–947

Zhang, F., Zhang, Y.F., and Nee, A.Y.C., 1997 Using

genetic algorithms in process planning for job shop

machining IEEE Transactions on Evolutionary

Computa-tion, 1 (4), 278–289

Zhang, F.P., 2008 Research on graph theory-based facturing setup planning Journal of Advanced Manufac-turing Systems, 7 (2), 313–318

manu-Zhang, H.C and Lin, E., 1999a A hybrid-graph approachfor automated setup planning in CAPP Robotics andComputer-Integrated Manufacturing, 15 (1), 89–100.Zhang, S., 2002 A web-based expert system on computeraided process planning In: International Conference onControl and Automation (ICCA), 16–19 June 2002214–214

Zhang, W.J and Xie, S.Q., 2007 Agent technology forcollaborative process planning: a review InternationalJournal of Advanced Manufacturing Technology, 32 (3–4),315–325

Zhang, X.Q., et al., 2006 Feature based integrated intelligentsequence design for cold extrusion Journal of MaterialsProcessing Technology, 174 (1–3), 74–81

Zhang, Y., et al., 1999a Object oriented manufacturingresource modelling for adaptive process planning Inter-national Journal of Production Research, 37 (18), 4179–4195

Zhang, Y.F., Ma, G.H., and Nee, A.Y.C., 1999b Modelingprocess planning problems in an optimization perspec-tive Proceedings - IEEE International Conference onRobotics and Automation, 3, 1764–1769

Zhao, F., et al., 2004 A genetic algorithm based approachfor integration of process planning and productionscheduling In: Proceedings - 2004 International Con-ference on Intelligent Mechatronics and Automation,Chengdu483–488

Zhao, F and Wu, P.S.Y., 1999 A cooperative framework forprocess planning International Journal of ComputerIntegrated Manufacturing, 12 (2), 168–178

Zhao, F.L., Tso, S.K., and Wu, P.S.Y., 2000 Cooperativeagent modelling approach for process planning Compu-ters in Industry, 41 (1), 83–97

Zhao, Y., Ridgway, K., and Al-Ahmari, A.M.A., 2002.Integration of CAD and a cutting tool selection system.Computers and Industrial Engineering, 42 (1), 17–34.Zhao, Y.F., Habeeb, S., and Xu, X., 2009 Research intointegrated design and manufacturing based on STEP.International Journal of Advanced Manufacturing Tech-nology, 44 (5–6), 606–624

Zhou, X., Du, P., and Liao, W., 2007a A feature andknowledge based intelligent CAPP method for auto-bodyspot-welding China Mechanical Engineering, 18 (13),1560–1565

Zhou, X., et al., 2007b A feasible approach to theintegration of CAD and CAPP CAD Computer AidedDesign, 39 (4), 324–338

A new nonlinear model for multiple criteria supplier-selection problem

An illustrative example is presented to compare the proposed model and those in the literature

Keywords: supplier selection; multiple criteria analysis; nonlinear programming

1 Introduction

In today’s highly competitive environment, an effective

supplier-selection process is very important to the

success of any manufacturing organisation and

select-ing the right supplier is always a difficult task (Liu and

Hai 2005, Sevkli et al 2007) The success of a supply

chain is highly dependent on selection of good

suppliers Supplier selection and evaluation is the

process of finding the appropriate suppliers who are

able to provide the buyer with the right quality

products and/or services at the right price, in the

right quantities and at the right time (Mandal and

Deshmukh 1994, Zhang et al 2009) Indeed supplier

selection is a multiple criteria decision-making problem

affected by several conflicting factors such as price,

quality and delivery (Gunery et al 2008) Early in

1960s, Dickson identified 23 criteria that ought to be

considered by personnel in evaluating suppliers

(Dickson 1966)

Over the years, several techniques have been

developed to solve the problem efficiently Analytic

hierarchy process, analytic network process, linear

programming, mathematical programming,

multi-ob-jective programming, neural networks, case-based

reasoning, simple multi-attribute technique and fuzzy

set theory methods have been applied in literature

(Chan 2003, Humphreys et al 2003a,b, 2006, Chan

and Chan 2004, Choy et al 2004, Ding et al 2005,

Chan and Kumar 2007, Chan et al 2007, 2008, Sevkli

et al 2007, Gunery et al 2008, Bachlaus et al 2009,

Guneri and Kuzu 2009) These models provide

systematic approaches for purchasing managers to

evaluate and score suppliers with multi-criteria theless, these models are not easy to implement Forinstance, models based on multi-objective optimisationrequire the decision makers to exogenously specify theexact values of weights of individual criteria It is,however, difficult to obtain precise weight values (Ni

Never-et al 2007, Ng 2008)

In a recent article, Ng (2008) proposed a weightedlinear optimisation model for multi-criteria supplier-selection problem The proposed model, hereaftercalled the Ng-model, converts all criteria measures of

a supplier into a scalar score The selection based onthe calculated scores is then done With propertransformation, the Ng-model can obtain the scores

of suppliers without a linear optimiser The Ng-model

is simple and easy to understand Despite its manyadvantages, the Ng-model leads to a situation wherethe weight of a certain criterion becomes zero That is,this criterion does not have any role for determiningtotal score of the related supplier This may lead to asituation where a supplier is inappropriately ranked.This may not reflect the real position of this supplier.The purpose of this article is to present an extendedversion of the Ng-model by considering weights valuesfor multi-criteria supplier-selection problem

The rest of this article is organised as follows Inthe following section, the Ng-model would be re-viewed In this section, shortcomings of the Ng-model

is also shown In Section 3, the proposed model issuggested Section 4 is devoted to a numericalillustration Section 5 is devoted to a real case study.Finally, conclusions and future research directions arepresented

*Email: ahadi@khuisf.ac.ir

Vol 24, No 1, January 2011, 32–39

ISSN 0951-192X print/ISSN 1362-3052 online

Ó 2011 Taylor & Francis

DOI: 10.1080/0951192X.2010.527372

2 Ng-model

Assume that I suppliers are available for a company

The purchasing manager would like to evaluate these

suppliers based on J criteria In particular, let the

performance of ith supplier in terms of each of the

criterion, j, be denoted as xij For simplicity, further

assume that all the criteria are benefit-type criteria, i.e

they are positively related to the score of a supplier If

there is a negatively related criterion, transformation

of negativity or taking reciprocal can be applied for

conversions The purpose is to aggregate multiple

performance scores of a supplier with respect to

different criteria into a single score Si In the

Ng-model, the author first transforms all measures to

comparable base Using transformation

yij¼ xij mini¼1; ;Ifxijg

maxi¼1; ;Ifxijg  mini¼1; ;Ifxijg ð1Þ

Ng converts all measurement in a 0–1 scale for all items

To facilitate the supplier selection under multiple

criteria, Ng defines a non-negative weight wijthat is the

weight of contribution of performance of the ith supplier

under the jth criteria to the score of the supplier It is

assumed that the criteria are ranked in a descending

order such that wi1 wi2    wiJfor all supplier i

The purpose is to aggregate multiple performance scores

of a supplier with respect to different criteria into a single

score The proposed model by Ng (2008) for aggregation

jwij¼ 1

uij 0; j¼ 1; ; J

ð3Þ

Now the maximal score Sican be obtained by the dual

of (3) That is, the score Si of the ith supplier can be

easily obtained as maxj¼1; ; Jð1

j

Pj k¼1yikÞ

2.1 Issues on Ng-modelHere using a multi-criteria supplier-selection problem, itwould be shown that Ng-model is not appropriate toapplications Three criteria are under consideration by acompany There are five suppliers available Themeasures of each supplier under the three criteria arelisted in Table 1 We take a reciprocal transformation ofthe second criterion so that the transformed values arepositively related to the desired scores Normalisation isthen performed to scale all measures within a 0–1 range.Table 2 shows the transformed and normalised measures

of all suppliers

Now the Ng-model [the model (2)] is applied tosolve this supplier-selection problem The followingtable shows the score of each supplier and optimalweight for each criterion

Table 3 shows the obtained results using the model As for the third criterion for all suppliers, theweight is 0, which means that the third criterion doesnot have any meaning Besides, the second criterion isconsidered only for two suppliers Therefore, we cansay that the Ng-model is not appropriate or applicable

Ng-3 Proposed model

In virtue of its data envelopment analysis (DEA) feature,the Ng-model avoids subjectiveness in determiningweights and provides an objective way for supplier-

Table 1 Measures of suppliers under criteria

Supplier 1st criterion 2nd criterion 3rd criterion

Table 2 Transformed and normalised measures

Supplier 1st criterion 2nd criterion 3rd criterion

selection problem However, as it has been observed, the

Ng-model may ignore the data of a criterion This may

lead to the situation where a supplier is inappropriately

ranked, which may not reflect the real position of this

supplier To address this issue, an extension of the

Ng-model is presented and proposes a similar weighted

optimisation model A situation in which a set of

suppliers is available is considered The manager would

like to rank these suppliers based on J criteria The

measure of supplier i under criterion j is denoted as xij

We evaluate a supplier by converting multiple measures

under all criteria into a single score A common scale for

all measures is also an important issue A particular

criterion measure, in a large scale, may always dominate

the score For this, all measures xijare normalised into a

0–1 scale All transformed measures are denoted as yij

To transform the performance ratings, the performance

ratings are normalised into the range of [1] by the

following equations (Cheng 1999)

(1) The larger the better type:

The score of a supplier is expressed as the weighted

sum of transformed measures Now let wj be the

relative importance weight attached to the jth criterion

and yij be the performance of ith supplier in terms of

jth criterion The proposed model is as follows:

The model (6) is a nonlinear programming model,

which determines the most favourable weights for each

supplier The model (6) is a variant of the following

multiple attribute decision-making model:

max Si¼XJ

j¼1

sijwj

s:t:XJ J¼1

wj ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiPsij

J j¼1s2 ij

However, due to the presence of the orderingconstraint w1 w2     wJ 0, the model (6)cannot usually be solved analytically, but can besolved using Microsoft Excel Solver or the LINGOsoftware package very easily

3.1 Sensitivity analysisThe model is simple-to-understand and easy-to-use.However, exogenous specification of ranking ofcriteria is required The results may be dependent onthe sequence of user-defined ranking One can examinethe sensitivity of change to a supplier score if criteriaranking changed In addition to change in rankingorder, any change in normalisation method maychange the supplier score

4 Numerical illustrations

In this section, an implementation of the proposedmodel is illustrated with a multi-criteria supplier-selection problem as in the literature (Liu et al 2000,

Ng 2008) Five criteria, including supply variety,quality, distance, delivery and price are under con-sideration by a firm manufacturing agricultural andconstruction equipment Supply variety is the number

of parts supplied by the supplier It is considered first

as the company would like to reduce the number ofsuppliers The quality of supplied parts is also animportant criterion for a company in supplier evalua-tion The distance is related to delivery efficiency Alonger distance will affect the delivery service of thesupplier due to a longer lead time or restricted deliverytime windows The criterion delivery measures thepercentage of on-time delivery Finally, the price indexindicates the estimated price level offered by a supplier

as compared to the average market price If the pricelevel offered is higher than the average price, the priceindex will be of a value higher than 100% and viceversa

There are 18 suppliers available The measures ofeach supplier under the five criteria are listed inTable 4 A reciprocal transformation of price anddistance measures is chosen for the transformed valuesare positively related to the desired scores Normal-isation is then performed to scale all measures within a

0–1 range Table 5 shows the transformed and

normalised measures of all suppliers

Using the Ng-model, Table 6 shows the obtained

weights for each supplier and its score (rank)

Evidently, the Ng-model does not consider the last

criterion for all suppliers and the fourth criterion is

considered only for three suppliers; in fact its

emphasise is on the first and second criterion Besides,

the Ng-model could not assist the manager in

obtaining a preferable and robust ranking result for

suppliers (see the score of suppliers 3 and 4) Table 7

shows the supplier selection using proposed model.This table shows the rank of each supplier in theproposed model, Ng-model (Ng 2008) and DEA model(Liu et al 2000) as well

For comparison purpose, the best five suppliers isconsidered as there were five efficient suppliersidentified by the Ng-model in Ng (2008), using thesame dataset The top five suppliers identified aresuppliers 15, 17, 10, 5 and 11 These suppliers are goodsuppliers in the Ng-model as well, but with differentranking In fact, the top five suppliers in the Ng-modelare 10, 17, 15, 5 and 11 As it is obvious, suppliers 17and 11 have the same rank in both Ng and theproposed model It can be seen from Table 7 thatsupplier 15 has the first rank in the proposed modelwhereas its rank in the Ng-model is 3 The reason isthat our model considers all of the five criteria whilethe Ng-model considers only the first and secondcriteria, that is, the weight of the third, fourth and fifthcriteria is zero in the Ng-model Now consider supplier

10 This supplier has the first rank in the Ng-modelwhile the rank of this supplier in the proposed model isthree To explain this difference, note that according toTable 6 the Ng-model only considers the firstcriterion (w1¼ 1) and ignores the other criteria (w2¼

w3¼ w4¼ w5¼ 0); while according to Table 7 theproposed method considers all of the criteria Further-more, the proposed model provides a robust rankingwhile as it has been noticed, the Ng-model does nothave this property

The above example has also been solved (usingDEA) in Liu et al (2000) For comparison purpose,the best five suppliers is again considered as there werefive efficient suppliers identified by the DEA model inLiu et al (2000) The top five suppliers identified are

Table 4 Measures of suppliers under criteria

Distance(Mile)

Delivery(%)

Priceindex(%)

Table 6 Obtained results using Ng-model

Supplier w1 w2 w3 w4 w5 Score Rank

suppliers 10, 17, 5, 15 and 11 Suppliers 10, 15 and 17

are good suppliers in both DEA and the proposed

model Suppliers 5 and 11 were not identified as good

suppliers in the DEA model On the other hand,

suppliers 1 and 12 were identified as good suppliers in

the DEA model but were not identified by the

proposed model The reasons for these differences are

due to the incorporation of the relative importance of

the criteria Suppliers 1 and 12 were efficient suppliers

in DEA models However, the supply varieties of these

two suppliers are only 2 and 7, which are relatively

low, compared to other suppliers When the supply

variety is considered as a relatively important criterion,

these two suppliers are eliminated The good suppliers

identified in the proposed model are good not simply

by the most important criterion (supply variety)

Suppliers 5 and 11 with relatively low supply variety

measures, 24 and 10, respectively, were rated high

because of the advantage of relatively shorter

dis-tances Finally note that similar to the Ng-model the

DEA model proposed in Liu et al (2000) could not

rank the suppliers Hence, the proposed model

there-fore provides a more reasonable and encompassing

index for supplier-selection problem as compared to

the Ng-model and DEA model

5 Case study

In this section, a case study is presented to better

describe the model The case study is related to the

supplier selection of the Energy Company (EC) EC

concentrates on producing solar boiler and solar water

refiners in Iran This company, to produce its products,

is required to purchase solar panels with different sizes

and voltages Hence, Energy Company buys its solarpanels from different suppliers with respect to its type

of home and industrial customers

At present 10 potential suppliers have beenidentified, all with strong reputations in at least onearea EC has identified the five rights as fundamentalcriteria for selecting suppliers Based on past dealingswith these 10 firms, as well as reliable documentationfrom those firms, EC has calculated the average priceper unit for items in the commodity group for each ofthe suppliers EC has also examined responsiveness ofthose firms and has noted the typical lead time foreach No particular objective measures were availablefor evaluating the firms according to the other criteria,but subjective evaluation is possible based on thedocumentation provided by the firms As a result,the procurement manager has been able to study thedocumentation and rate each firm according to aseven-point Likert-type scale on each of the subjectivecriteria More typically, this type of subjective evalua-tion would be the result of the work of a cross-enterprise commodity team In such cases, eachmember of the team would perform the subjectiverating task and the evaluations of each member would

be averaged for each criterion These averages wouldthen be used in place of the manager’s assessments.Either way, a set of raw scores could be assembled foreach of the vendors being considered

Table 8 summarises the raw score data for the 10firms being considered Values indicated for price aregiven as average cost per unit for the items beingconsidered For lead time, the values given are in days.Thus, for both price and lead time, small data valuesare preferable The subjective ratings for the other

Table 7 Obtained results using proposed model and its comparison with Ng and DEA model

RankProposed model Ng-model DEA

criteria are such that a value of 7 indicates the best

performance that might be expected and a value of 1

indicates the worst conceivable performance Note that

each supplier other than supplier 2 is ‘‘best’’ at

something; e.g supplier 1 has the best quality (tied

with suppliers 5 and 6), while supplier 3 rates best with

respect to its ability to deliver directly to the right

place, supplier 7 has the best price, supplier 6 has the

best lead time, and so forth Table 9 shows the

transformed suppliers data for EC This table shows

the score and the rank of each supplier using the

proposed model, too As it seems, the best three

suppliers are S1, S10 and S6 The author solved EC

supplier-selection problem by Ng-model Table 10

shows the obtained results using Ng-model Table 10

shows that the ranking by Ng-model is completely

different from proposed model Tables 8 and 9 show

that only S4 has the same rank in both methods

Suppose the company would like to swap the order

of importance of criteria quality and price The new

supplier scores and ranks are given in Table 11 under

the revised criteria ranking

5.1 Linear discriminant analysis

Now a statistical method is presented to justify

obtained results Linear programming methods of

discriminant analysis have received a great deal of

attention recently, e.g see Glover (1990), Ragsdale andStam (1991), Glover et al (1988), among others.Linear discriminant analysis (LDA) is a method fordetermining group classification for a set of similarunits or observations For example, LDA may be used

to classify loan applicants as either good or bad creditrisks, and hence determine whether to accept or rejectnew applicants The same relevant factors are mea-sured for all units in the set and there is some distinctoutcome or occurrence, which determines groupmembership for each unit, e.g whether the loanrecipient defaults on the loan The objective of anLDA is to find a set of factor weights which bestseparates the groups, given a set of units for whichgroup membership is already known The resulting set

of weights may then be used to predict groupmembership for new units

LDA and the related Fisher’s linear discriminant aremethods used in statistics, pattern recognition andmachine learning to find a linear combination offeatures, which characterise or separate two or moreclasses of objects or events LDA is closely related toANOVA (analysis of variance) and regression analysis,which also attempt to express one dependent variable as

a linear combination of other features or measurements(Fisher 1963, McLachlan 2004) In the other twomethods, however, the dependent variable is a numericalquantity, while for LDA it is a categorical variable (i.e.the class label) Logistic regression and probit regressionare more similar to LDA, as they also explain acategorical variable LDA is also closely related toprincipal component analysis (PCA) and factor analyses

in that both look for linear combinations of variableswhich best explain the data (Martinez and Kak 2004).LDA explicitly attempts to model the difference betweenthe classes of data PCA on the other hand does not takeinto account any difference in class, and factor analysisbuilds the feature combinations based on differencesrather than similarities

In general, LDA involves two or more groups, butthe two-group scenario has received the most atten-tion, as in Glover (1990) This article considers the

Table 9 Transformed suppliers data for EC

Supplier Price (USD) Quality Lead time Quantity Delivery Score Rank

two-group case as in literature The author divides 10

suppliers into two groups In the first group, there are

three suppliers, which were considered as best one in

the proposed model, i.e S1, S10 and S6 And, the

remaining suppliers are in the second group To survey

analyse the data, SPSS software is used By LDA, the

following results were observed: Table 12 shows the

correlation coefficient of the criteria As it appears, no

criteria have the significant correlation coefficient The

special values are shown in the Table 13, in which the

focus correlation is 0.929 According to the Table 14,

the performed categorisation is fully significant,

because it is Sig ¼ 0.033 It is necessary Sig 5 0.05

to be significant Table 15 shows that all three suppliers

dedicated to the first group by the proposed model

were confirmed to be in the first group by LDA

Finally, as Table 15 shows the initial classification by

the proposed model is confirmed by LDA for 100% of

the suppliers

6 Summary and future directionsThe issues of supplier selection have attracted theinterest of researchers since the 1960s, and researches

in this area have evolved Perhaps the greatestsignificance of any good multicriteria procedure isthat it provides a structure to guide the decision makerthrough a complex decision process, such as supplierselection Criteria must be identified and consideredsystematically, as must alternatives (e.g suppliers).The decision maker is forced to be thorough in theassessment of the problem and in the evaluation of thealternatives available This article presented a simplenonlinear programming model for multi-criteria sup-plier-selection problem The contribution of this article

is to provide a model for supplier-selection problemthat not only incorporates multiple criteria but alsomaintains the effects of weights in the final solution.Limitations of the proposed model are as follows:

(1) Being a deterministic rather than statisticaltechnique, the proposed model produces resultsthat are particularly sensitive to measurementerrors

(2) Since the proposed approach creates a linear program for each supplier, large pro-blems can be computationally intensive

non-For future research, these two aspects, selection criteria and methods, will continue to be thefocus For supplier-selection criteria, combining supply

supplier-Table 10 Suppliers ranks by Ng-model

Table 11 Suppliers data for EC

Table 12 Pooled within groups matrices

Correlation Price Quality

Leadtime Quantity Delivery

Cumulative

%

Canonicalcorrelation

Table 14 Wilks’ lambda

Test of function Wilks’ lambda Chi-square df Sig

Table 15 Classification results

Rank

Predicted groupmembership

chain performance measurement and supplier selection

seems to be an important area Although some articles

are on supply chain management environment, little

attention has been paid on the influences on the whole

supply chain if a certain supplier is selected Some new

criteria to reflect the whole supply chain performance

should be developed in the process of supplier

selection The method mentioned in this study has

shortcomings in dealing with the selection problem

New methods to simulate the process of human

decision-making, such as neural network, seem to be

promising, and the computer programming for

suppli-er selection should also be developed

Finally, while the supplier-selection problem has been

used to illustrate a decision support application for the

proposed model, a variety of other applications could

also serve as areas for the similar use of the proposed

model In particular, with the increased emphasis lately

on improved risk analysis and project management, our

model could readily be applied to decisions such as

project selection, personal selection and so on

Acknowledgements

Appreciation is expressed to M Mirjaberi and A Yousefi

for their help The insightful comments from the anonymous

reviewers are sincerely appreciated Their constructive

suggestions on the earlier versions have significantly

im-proved this paper

References

Bachlaus, M., Tiwari, M.K., and Chan, F.T.S., 2009

Multi-objective resource assignment problem in a

product-driven supply chain using a Taguchi-based DNA

algorithm International Journal of Production Research,

47, 2345–2371

Chan, F.T.S., et al., 2008 Global supplier selection: a

fuzzy-AHP approach International Journal of Production

Research, 46, 3825–3857

Chan, F.T.S and Chan, H.K., 2004 Development of the

supplier selection model – A case study in the advanced

technology industry Proceedings of the Institution of

Mechanical Engineers Part B: Journal of Engineering

Manufacture, 218 (12), 1807–1824

Chan, F.T.S., et al., 2007 A decision support system for

supplier selection in the airline industry Proceedings of

the Institution of Mechanical Engineers Part B: Journal of

Engineering Manufacture, 221, 741–758

Chan, F.T.S and Kumar, N., 2007 Global supplier

development considering risk factors using fuzzy

ex-tended AHP-based approach Omega-International

Jour-nal of Management Science, 35 (4), 417–431

Chan, F.T.S., 2003 Interactive selection model for supplier

selection process: an analytical hierarchy process

ap-proach International Journal of Production Research, 41

(15), 3549–3579

Choy, K.L., et al., 2004 Design of an intelligent supplier

relationship management system for new product

devel-opment International Journal of Computer Integrated

Manufacturing, 17 (8), 692–715

Dickson, G.W., 1966 An analysis of vendor selection

systems and decisions Journal of Purchasing, 2 (1), 5–17

Ding, H., Benyoucef, L., and Xie, X., 2005 A simulationoptimization methodology for supplier selection pro-blem International Journal Computer Integrated Manu-facturing, 18 (2–3), 210–224

Fisher, R., 1936 The use of multiple measurements intaxonomic problems Annals of Eugenics, 7, 179–188.Glover, S., Keene, S., and Duea, B., 1988 A new class ofmodels for the discriminant problem Decision Sciences,

19, 269–280

Glover, F., 1990 Improved linear programming models fordiscriminant analysis Decision Sciences, 21, 771–785.Gunery, A.F., Yucel, A., and Ayyildiz, G., 2008 Anintegrated fuzzy-LP approach for a supplier selectionproblem in supply chain management Expert Systemwith Application, 36, 9223–9228

Guneri, A.F and Kuzu, A., 2009 Supplier selection by using

a fuzzy approach in just-in-time: a case study tional Journal of Computer Integrated Manufacturing, 22,774–783

Interna-Humphreys, P., et al., 2006 Employing dynamic fuzzymembership functions to assess environmental perfor-mance in the supplier selection process InternationalJournal of Production Research, 44, 2379–2419

Humphreys, P.K., McIvor, R., and Chan, F.T.S., 2003a.Using case-based reasoning to evaluate supplier environ-mental management performance Expert Systems withApplications, 25 (2), 141–153

Humphreys, P.K., Wong, Y.K., and Chan, F.T.S., 2003b.Integrating environmental criteria into the supplierselection process Journal of Materials Processing Tech-nology, 138 (1–3), 349–356

Liu, F., Ding, F.Y., and Lall, V., 2000 Using dataenvelopment analysis to compare suppliers for supplierselection and performance improvement Supply ChainManagement, 5 (3), 143–150

Liu, F.H.F and Hai, H.L., 2005 The voting analytichierarchy process method for selecting supplier Interna-tional Journal of Production Economics, 97, 308–317.McLachlan, H., 2004 Discriminant analysis and statisticalpattern recognition New York: Wiley Interscience.Mandal, A and Deshmukh, S.G., 1994 Vendor selectionusing interpretive structural modelling (ISM) Interna-tional Journal Operations and Production Management,

14 (6), 52–59

Martinez, B and Kak, S., 2004 PCA versus LDA IEEETransactions on Pattern Analysis and Machine Intelli-gence, 23 (2), 228–233

Ng, W.L., 2008 An efficient and simple model for multiplecriteria supplier selection problem European Journal ofOperational Research, 186, 1059–1067

Ni, M., Xu, X., and Deng, S., 2007 Extended QFD anddata-mining-based methods for supplier selection in masscustomization International Journal of Computer Inte-grated Manufacturing, 20, 280–291

Ragsdale, C.T and Stam, A., 1991 Mathematical ming formulations for the discriminant problem: an olddog does new tricks Decision Sciences, 22, 296–307.Sevkli, M., et al., 2007 An application of data envelopmentanalytic hierarchy process for supplier selection: a casestudy of BEKO in Turkey International Journal ofProduction Research, 45, 1973–2003

program-Zhang, D., et al., 2009 An novel approach to supplierselection based on vague sets group decision ExpertSystem with Applications, 36, 9557–9563