International journal of computer integrated manufacturing , tập 24, số 6, 2011

Integration of process planning and scheduling: a state-of-the-art reviewRakesh Kumar Phanden, Ajai Jain* and Rajiv VermaDepartment of Mechanical Engineering, National Institute of Techn

Integration of process planning and scheduling: a state-of-the-art review

Rakesh Kumar Phanden, Ajai Jain* and Rajiv VermaDepartment of Mechanical Engineering, National Institute of Technology, Kurukshetra, India

(Received 12 July 2010; final version received 6 February 2011)Process planning and scheduling functions strongly influence profitability of manufacturing a product, resourceutilisation and product delivery time Several researchers have addressed the need for integration of process planningand scheduling (IPPS) functions to facilitate flexibility and for improving profitability of manufacturing a product,delivery time as well as creation of realistic process plans that can be executed readily on shop floor This articlepresents a state-of-the-art review on IPPS Three common integration approaches, non-linear approach, closed loopapproach and distributed approach, are discussed with their relative advantages and disadvantages and reportedresearch is classified accordingly It also identifies several future research directions

Keywords: integration; process planning; scheduling; review

1 Introduction

Process planning and scheduling are two most

important tasks in a manufacturing system These

tasks strongly influence profitability of manufacturing

a product, resource utilisation and product delivery

time (Yang et al 2001) Process planning is the

systematic determination of methods by which a

product is to be manufactured economically and

competitively The primary goal of process planning

function is to generate process plans, which specifies

raw material/components needed to produce a product

as well as processes and operations necessary to

transform raw materials into the final product Thus,

outcome of process planning is the information

required for manufacturing processes, including

iden-tification of machines, tools and fixtures Scheduling

assigns a specific task to a specific machine in order to

satisfy a given performance measure It is bound by

process sequencing instructions that the process plan

dictate and by the time-phased availability of

produc-tion resources Thus, both process planning and

scheduling involve assignment of resources and are

highly interrelated Conventionally, process planning

and scheduling are carried out in two distinct,

sequential phases, where scheduling is done separately,

after the process planning This approach is based on

the concept of subdividing the tasks into smaller and

separated duties to satisfy the requirements of

sub-optimisation and suitable for mass production (Larsen

and Alting 1992) However, today’s manufacturing

environment is quite different from traditional one It

is characterised by decreasing lead time, exactingstandards of quality, larger part variety and competi-tive costs In such manufacturing environment, it isdifficult to get a satisfactory result using traditionalapproach due to following reasons: (Larsen and Alting

1992, Zhang and Merchant 1993, Gindy et al 1999,Morad and Zalzala 1999, Baykasoglu and O¨zbakır

2009, Li et al 2010a,b,c)

(1) Process planner assumes that shop floor is idleand unlimited capacities of resources are al-ways available in the shop Thus, processplanner plans for the most recommendedalternative resources This leads to the processplanner favouring to select the desirable re-sources repeatedly Moreover, the resources arenever always available on shop floor There-fore, unrealistic process plan will generate thatmay not be readily executed on shop floor.(2) In conventional approach, fixed process plansrestrict the schedule to only one machine peroperation Therefore, possible choices of sche-dule using alternative machines are ignored.(3) Even if the dynamic shop status is consideredduring process planning phase, the constraintsconsidered in planning phase may changegreatly because of time delay between planningphase and scheduling phase Thus, the gener-ated process plan may become sub-optimal orinvalid

(4) Both, process planning and scheduling focus

on single criterion optimisation to determine

*Corresponding author Email: ajayjainfme@nitkkr.ac.in

International Journal of Computer Integrated Manufacturing

Vol 24, No 6, June 2011, 517–534

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

Ó 2011 Taylor & Francis

DOI: 10.1080/0951192X.2011.562543

optimal solution However, real manufacturing

environment involves more than one

optimisa-tion criterion

The short comings of traditional approach can be

overcome by considering an integrated approach to

process planning and scheduling An integrated

approach can respond better than traditional approach

to present manufacturing environment and facilitate

flexibility, improves profitability of a product, resource

utilisation, product delivery time and creation of

realistic process plans that can readily be executed

without frequent alterations (Chryssolouris and Chen

1985, Sundaram and Fu 1988, Saygin and Kilic 1999,

Lee and Kim 2001, Kumar and Rajotia 2003) Thus,

integration of process planning and scheduling (IPPS)

is essential to achieve eventually integrated

manufac-turing and to dismiss conventional manufacmanufac-turing

approach

The purpose of this article is to present a

state-of-the-art review in the area of IPPS by synthesis the

information available in literature The various

approaches for IPPS have been discussed with their

advantages and disadvantages and reported research is

classified accordingly It also identifies potential future

research directions Thus, this article not only provides

a platform to novice researchers but also assist in

stimulating further research in the area of IPPS This

article is organised with the following sections Section

2 presents the various approaches to integration along

with related contributions Section 2.1 discusses

non-linear approach (NLA) Section 2.2 presents closed

loop approach (CLA) Section 2.3 deals with

distrib-uted approach (DA) Section 2.4 presents the other

approaches for IPPS that are followed by researchers

Section 3 presents the conclusion and shows some

potential future research directions

2 Literature review

The best way for IPPS is to merge both process

planning and scheduling functions into one However,

as process planning and scheduling individually are

non-polynomial (NP)-hard, the resulting problem is

also NP-hard (Khoshnevis and Chen 1990) Moreover,

process planning and scheduling department in a

company have to be completely dismantled and

reorganised Thus, it cannot be implemented in a

company with existing process planning and

schedul-ing departments Tan and Khoshnevis (2000)

at-tempted in this direction with a limited success

Another way of IPPS is to increase information

exchange between process planning and scheduling

functions Several classification schemes are suggested

by various researchers following this approach (Zhang

and Merchant 1993, Huang et al 1995, Gaalman et al

1999, Gindy et al 1999, Zhang et al 2003a, Shen et al

2006, Baykasoglu and O¨zbakır 2009, Guo et al 2009,Wang et al 2009, Li et al 2010b) However, presentwork follows the most commonly used classificationamong researchers (Larsen and Alting 1990, Zhang andMerchant 1993, Huang et al 1995, Gaalman et al 1999,Gindy et al 1999, Baykasoglu and O¨zbakır 2009).Accordingly, there are three main approaches of integra-tion viz., NLA, CLA and DA These IPPS approachesand related contributions are discussed below

2.1 Non-linear approachHere, multiple process plans (MPP) for each partbefore it enters to shop floor are created by consideringoperation flexibility (possibility of performing anoperation on more than a machine), sequencingflexibility (possibility of interchanging the sequence inwhich required manufacturing operations are per-formed) and processing flexibility (possibility ofproducing the same manufacturing feature withalternative operations or sequence of operations)(Benjaafar and Ramakrishnan 1996) The underlyingassumption is that all problems that can be solvedahead of time should be solved before the manufactur-ing starts Thus, NLA is based on static shop floorsituations (Zhang and Merchant 1993, Gaalman et al.1999) All these possible process plans are rankedaccording to process planning criterion (such as totalmachining time and total production time) and stored

in a process planning database The first priority plan

is always ready for submission when the job is requiredand then scheduling makes the real decision If the firstpriority plan does not fit well in the current status ofshop floor, the second priority plan is provided toscheduling This procedure is repeated until a suitableplan is identified from already generated process plans.The criteria for decisions are due dates and batch size

of order, capacity of workshop and optimisationcriterion for schedule (throughput, lead time, etc.).Figure 1 shows the NLA

NLA has one-way of information flow, i.e fromprocess planning to production planning, and thus, itmay be impossible to achieve full optimal results inintegrating the two functions (Kempenaers et al 1996).Moreover, modern production systems maintain MPP(Kim and Egbelu 1999), and it seems to be a proper

Figure 1 NLA (Zhang and Merchant 1993)

means to realise the integration between process

planning and scheduling (Kempenaers et al 1996)

Also, it can be implemented in a company with existing

process planning and scheduling department When

there are large numbers of parts, the number of process

plans tends to increase exponentially and can cause a

storage problem (Usher 2003) Also, some of the

process plans created are not feasible according to

real-time shop status and considering all possible process

alternatives for resource allocation may enormously

increase the complexity of process plan representation

(Zhang and Merchant 1993, Huang et al 1995)

Chryssolouris and Chan (1985) proposed

manu-facturing decision-making approach (MADEMA), the

first approach available in literature for IPPS It

considered a set of alternative resources for execution

of a particular production task Decision matrix was

formed for the selection of alternatives, where row

represents alternative while column represents

attri-bute and entry was value of attriattri-bute for

correspond-ing alternative MADEMA concept contained five

consecutive steps: (i) determine alternatives, (ii)

deter-mine attributes, (iii) deterdeter-mine consequences with

respect to attributes for each alternative, (iv) apply

decision rules for choosing the best alternative and (v)

select the best alternative The alternative resources

were chosen by evaluating the contribution on some

decision making established criterion such as a linear

combination of attributes with weights or alternatives

with greater chance to produce higher utility value

Sundaram and Fu (1988) developed a scheduling

method through outcome of process planning for

minimisation of makespan and to balance loads for

machines in job shop environment For schedule

improvement, authors used an automated system

based on group technology (GT) called integrated

computer-aided process planning and scheduling They

used a group scheduling algorithm called key machine

loading and combined it with process planning

generator and operation planner A key machine was

continuously

Tonshoff et al (1989) presented FlexPlan for IPPS

Authors created all MPP before manufacturing starts

The scheduling function selected the suitable process

plan according to the availability of resources This

approach covers reactive replanning to allow reaction

of disturbances occurring on shop floor

Srihari and Greene (1990) proposed a prototype

computer-aided process planning (CAPP) for a

Flex-ible Manufacturing System (FMS) to integrate

sche-duling function GT coding system was used to input

information of parts parameters Heuristic knowledge

was used to decide sequence of operations and route

through system, on basis of the flow time of jobs

A dynamic shop status module of prototype CAPPsystem maintained dynamic shop status overtime.Every alternative route was evaluated with respect tominimisation of flow time of jobs The queues at everymachine were modelled in pseudofacility that mon-itored in terms of time units Prismatic and rotationalparts were planned and tested with proposed ap-proach Authors concluded that proposed CAPPsystem for an FMS with multiple machining centresmaintained dynamic shop floor conditions in order todecide the actual sequence of operations and the finaljob route

Jablonski et al (1990) proposed a flexibly grated production planning and scheduling systemhaving three modules First was an automated featurerecognition module in order to identify geometricfeatures of a part and generate a production-orientedpart representation in term of basic manufacturingoperations Second was a process planning module(PPM) in order to generate all possible resourcescombinations for production of the part Third was ascheduling and dispatching module, which select thebest resources combination to produce the partaccording to some user defined strategy such asmanufacturing operations/features and resources onthe shop floor Authors showed that flexible andreactive scheduling approach on feature-based processplanning was feasible

inte-Palmer (1996) proposed a simulated annealing (SA)based approach for IPPS It contained three types ofconfiguration alterations; (i) reverse the order of twosequential operations on a machine, (ii) reverse theorder of two sequential operations within a job and(iii) change the method used to perform an operation.The cost functions such as, tardiness, mean flow time,makespan and a combined function of mean flow timeand tardiness were considered The performance of SAand dispatching rules were compared Authors con-cluded that solution quality of SA was remaining highacross varying situations, and it was effective meansfor IPPS Also, it outperformed the use of dispatchingrules

Kim and Egbelu (1999) proposed a mathematicalapproach to develop a scheduling tool for multiple jobswith each having MPP in a job shop environment.Authors claimed that the proposed methodologyminimise throughput/makespan for part mix It con-tained two sub-systems viz., process plan selectionsubsystem (PPSS) and shop scheduling subsystem(SSS) PPSS selected a set of process plans for eachpart type to be scheduled The selected set of processplans was passed to SSS to generate a feasibleschedule Performance measures determined by sche-duling system were passed back to process planningsystem to modify its process plan selection This

iterative process continues until no further

improve-ment in schedule was identified The scheduling

problem was solved with two algorithms viz.,

pre-processing algorithm and heuristic/iterative algorithm

‘Pre-processing algorithm’ combined features of

branch and bound and integer programming

techni-ques while heuristic/iterative algorithm combined

features of branch and bound technique and earliest

completion time dispatching rules They concluded

that computational time of pre-processing algorithm

was substantially lower than that of mix integer

programming technique but higher than that of

heuristic model As number of jobs increases, solution

quality obtained by heuristic got worse, but as the

number of machine increases, it had no clear effect on

performance of heuristics The increase in number of

process plans per job had a negative effect on the

solution quality of the heuristic

Aldakhilallah and Ramesh (1999) proposed an

architecture and framework called

computer-inte-grated process planning and scheduling (CIPPS) which

consists of three modules viz., super relation graph,

cover set model and cover set planning and scheduling

module (CSPS) in order to recognise polyhedral

depression features and extract prismatic features

from CAD database using artificial neural network

(ANN) and computational geometry techniques

Moreover, CIPPS framework contained three modes

of operations viz., dynamic support for design decision

(DSDD), runtime intelligent operational control (IOC)

and data consolidation and integration (DCI) DSDD

mode supports decisions during design process IOC

mode worked up for automatic shop floor

manage-ment, when changes occurred in the environment DCI

mode performed the interfacing and integration of

CIPPS with other functions in manufacturing

environment

Weintraub et al (1999) proposed a procedure for

scheduling jobs, to minimise manufacturing cost while

satisfying due dates by taking into account alternative

process plans of jobs, in a large scale manufacturing

job shop An iterative simulation-based scheduling

algorithm was developed to minimise lateness and

applied in virtual factory, which was a Windows-based

software package In order to further reduce lateness, a

Tabu search (TS)-based algorithm was applied to

identify process plans with alternative operations and

routings The numbers of alternative process plans

were fixed for the job at two Process plans with

alternative routings, operations and sequences were

selected according to the current shop status They

concluded that scheduling with alternatives can greatly

improve the ability to satisfy due dates under varying

shop status Also, scheduling with alternative

opera-tions had largest schedule improvement and schedule

with alternative sequence had smallest scheduleimprovement

Saygin and Kilic (1999) proposed a framework tointegrate MPP with predictive (off-line) scheduling in

an FMS, in order to minimise completion time Theframework worked up in four stages: (i) machine toolselection, (ii) process plan selection, (iii) schedulingand (iv) re-scheduling module Dissimilarity maximi-sation method (DMM) was used for selection ofappropriate process plans of a part mix Reschedulingstrategy was developed in order to reduce waitingtime of parts and algorithms were based on mathe-matical and heuristic approaches They concludedthat idle time of machine tool was reduced to 30 unitsfrom 81units and waiting time of parts was dropped

to 50 units from 74 units Also, optimal process planthat might have shortest processing time (SPT) orleast number of operations may not guarantee bestsystem performance

Lee and Kim (2001) proposed a method for IPPS,using simulation based on genetic algorithm (GA).Simulation module computes performance measuresbased on process plans combination created by GAinstead of process plan alternatives and output thenear-optimal process plan combination prior toexecution on shop floor The performance measureswere makespan and lateness based on SPT and earliestdue date (EDD) dispatching rules They concludedthat about 20% reduction of makespan was possiblewhen compared with random selection of process plancombination

Yang et al (2001) proposed a feature-based ple-alternative process planning system with schedul-ing verification The process plan was generateddirectly from part design and available resource datainformation The system had four components viz., arelational manufacturing database, form feature re-cognition, process alternative generation and schedul-ing state evaluation A 3D model and blank rawmaterial model was entered by using initial graphicsexchange specification data format The manufactur-ing features were decomposed by using graph-basedand rule-based algorithm After generating MPP, eachplan was allocated to scheduling, and a candidateprocess plan was retrieved on the basis of required duedate Authors concluded that proposed prototypecontained choice of process sequence with verification

multi-of delivery time for all feasible set multi-of process sequences.Moon et al (2002) proposed a GA-based IPPSmodel for multi-plant supply chain A mathematicalmodel was formulated with consideration of alter-native machines and sequences, sequences dependentsetup and due dates to minimise tardiness Operationssequencing problem was formulated as a multipletravelling salesman problem (TSP’s) and each TSP

determine machine operation sequences for each part

type A topological short technique (TST) was used to

obtain all flexible sequences in directed graph Authors

concluded that proposed GA approach was more

efficient than TS approach in terms of computational

time and problem size Also, population size and

number of generations were main factors that affect

performance of the proposed approach

Kim et al (2003) proposed an artificial intelligence

(AI) search technique called symbiotic evolutionary

algorithm (SEA) to simultaneously deal with process

planning and job shop scheduling in FMS SAE was

based on the fact that parallel searches for different

pieces of solution were more efficient than a single

search for the entire solution They considered

opera-tion flexibility, sequencing flexibility and processing

flexibility during process planning The job-shop

scheduling determines both process plan for each job

and corresponding schedule, while optimising two

types of objectives: minimising makespan and

mini-mising mean flow time SEA was tested on 24 test-bed

problem set and found better outcomes than existing

cooperative co-evolutionary GA (CCGA) (Potter

1997) as well as hierarchical approach

Kumar and Rajotia (2003) suggested a method for

on-line scheduling in a CAPP system for a job shop

environment A scheduling factor was used to make

operation-machine assignment The operations were

assigned to machines with highest value of actual

scheduling factor The scheduling criterions were flow

time and number of tardy jobs They concluded that

the proposed method helps in on-line determination

and assignment of loads on various machines Further,

Kumar and Rajotia (2006) proposed a framework for

IPPS system in job shop environment It considered

machine capacity and cost while assigning operation to

machines The proposed framework contained two

controlling modules viz., process plan generator and

scheduler Both modules were interacting with a

decision support system (DSS) DSS interacts with

various databases such as machine tool database,

tool-work material database and machining parameters

database A generative scheme was used to develop

process plan for axis-symmetric components The

scheduling factor as reported in earlier authors work

(Kumar and Rajotia 2003) was used to assign setup to

respective machine tool

Zhao et al (2004) proposed a GA-based approach

for IPPS in a job shop environment A fuzzy inference

system was used to select alternative machines It was

based on fuzzy logic toolbox by MATLAB Based on

the capability of machines, GA was used to balance

load for all machines Gliffer and Thompson’s

algo-rithm (Gliffer and Thompson 1960) was used to

evaluate fitness of chromosome in schedule builder

The scheduling objectives were to minimise makespan,minimise number of rejects and minimise processingcost Zhao et al (2006) extended their earlier work andused particle swarm optimisation (PSO) algorithm forbalancing load on each machine Moreover, Zhao et al.(2010) proposed an IPPS applicable to HolonicManufacturing System (HMS) in which they used ahybrid PSO and differential evolution algorithm inorder to balance the load for all machines

Grabowik et al (2005) proposed an integrationmethodology utilising MPP of a product in order torespond in disturbances during manufacturing Theproposed methodology represented a predictive-reac-tive and event-driven approach to rescheduling Theyconcluded that the availability of processes routesexpanded flexibility of control system and increasesefficiency of rescheduling Choi and Park (2006)proposed a GA-based method for IPPS, that minimisemakespan of each job order, considering alternativemachines and alternative operations sequences inintegrated manufacturing environment An opera-tion-based representation was used to constructchromosomes The performance of proposed methodwas evaluated in a job shop environment evolvingMPP Authors concluded that the proposed approachshows the possibility of improving makespan

Jain et al (2006) proposed an integration schemethat can take advantage of flexibility on the shop floorand can be implemented in a company with existingprocess planning and scheduling departments Theproposed methodology was able to take advantage ofMPP, while following a real-time strategy for schedul-ing suitable for changing workshop status Theproposed system was composed of two basic modules:process plans selection module (PPSM) and schedulingmodule (SM) PPSM selects best four process plans foreach part type and stores them in a database SMperforms part scheduling for using best four processplans The effectiveness of MPP over single processplan was assessed through makespan and mean flowtime Authors concluded that the availability of MPPduring FMS scheduling improves makespan and meanflow time

Li and McMahon (2007) proposed a SA-basedapproach for IPPS in a job shop environment Proces-sing, operation sequencing and scheduling flexibilitywere used to explore search space of proposed algo-rithm The algorithm was defined in two sets of datastructures The first set represents process plans and thesecond set specifies the schedule of a group of parts Theperformance measures were makespan, balanced level

of machine utilisation, job tardiness and manufacturingcost The proposed algorithm was compared with GA,

TS and PSO algorithms The authors concluded that theproposed algorithm performed satisfactorily and wasInternational Journal of Computer Integrated Manufacturing 521

able to choose one or more specific performance

criterion to address practical requirements

Moon et al (2008) proposed an evolutionary search

method based on TST for IPPS in supply chain A

mixed integer programming model was formulated,

which incorporate process planning of resources

selec-tion and sequence of operaselec-tion as well as determinaselec-tion

of their schedule to optimise makespan They

con-ducted three experiments with considerations of various

orders, operations and resource sizes and concluded

that proposed approach was capable of producing

optimal schedule and robust in generating the best

makespan through varied genetic environment and

under various order environments with precedence

constraints

Li et al (2008b) proposed a GA-based approach to

facilitate IPPS They developed an efficient genetic

representation and operator scheme The first part of

chromosomes composed of alternative process plan

string and second part composed of scheduling plan

string They assumed job shop environment to minimise

makespan They found that the value of makespan

without integration was worse than proposed

integra-tion model

Li et al (2010c) proposed a hybrid approach, which

synthesises advantage of GA and TS to solve IPPS

problem The first part of chromosome was alternative

process plan string, second part was scheduling plan

string and third was machine string Third part selects

machine set of corresponding operations of all jobs to

minimise makespan Authors concluded that the

proposed algorithm was effective and acceptable for

IPPS problem

Wang et al (2008) proposed an IPPS approach in a

batch manufacturing environment by utilising process

plan solution space A heuristic was developed to

minimise tardiness and also, to maintain cost of

process plan involved in modification of process

plan An SA algorithm was used to find optimal

process plan for prismatic parts only They concluded

that tardiness of jobs was improved, while the cost

of process plan was maintained at low level, because

PPM optimises the route with minimum processing

cost

Haddadzade et al (2009) proposed an approach for

IPPS, in a job shop for prismatic components that can

be implemented in a company with existing

depart-ments The model consisted of PPM and SM PPM

generates all possible alternative plans, then SM

ranked these based on minimum cost while due date

was considered The proposed approach took

advan-tage of MPP to fulfil due dates using overtime It can

optimise cutting parameters for milling operations

only Authors concluded that the proposed approach

can determine machining parameter, tool, machine

and amount of overtime within minimum costobjective and due date

Baykasoglu and O¨zbakır (2009) proposed an IPPSmodel that comprises of two parts First part was ageneric process plan (GPP) generator to generate finalprocess plan Second part was dispatching rule basedheuristic to generate feasible schedules A multipleobjective tabu search algorithm was employed to find

an optimal schedule for two objectives that were ‘flowtime’ and ‘cost of process plan’ They concluded thatprocess plan cost decreases as process plan flexibilityincreases

Rajkumar et al (2010) proposed a multi-objectivegreedy randomised adaptive search procedures(GRASP) in order to minimise makespan, maximumworkload, total workload, tardiness and total flow timeevolving flexible job shop environment It consists oftwo phases viz., construction phase and local searchphase Authors focussed on construction phasethrough computational experiments to solve IPPSproblem The IPPS framework consisted of PPSMand SM The proposed algorithm was validated withfour benchmarking problems Authors concluded thatthe proposed GRASP was effective to solve IPPSproblem

Leung et al (2010) proposed an IPPS approachutilising ant colony optimisation (ACO) algorithmbased on multi-agents system (MAS) in order tominimise makespan evolving job shop environment.They considered processing flexibility of alternativerouting and alternative machines AND/OR graphswere used to represent MPP Authors concluded thatthe proposed agent-based ACO approach was feasible

to solve IPPS problem

2.2 Closed loop approachHere, process plans are generated by means of adynamic feedback from production scheduling andavailable resources Production scheduling tells processplanning regarding availability of different machines

on shop floor for the coming job, so that every plan isfeasible with respect to current availability of produc-tion facilities Every time an operation is completed onshop floor, a feature-based work piece description isstudied in order to determine next operation andallocate the resources This approach takes dynamicbehaviour of the manufacturing system into considera-tion Thus, real-time status is crucial for CLA (Zhangand Merchant 1993) It is also referred to as real-timeapproachor dynamic approach Figure 2 shows CLA

In order to take full advantage of CLA, processplanning and scheduling departments in a companymay have to be dismantled and reorganised (Iwata andFukuda 1989) Moreover, it requires high-capacity

software and hardware (Zhang and Merchant 1993)

and adaptation of step-by-step local view that limits

the solution space for subsequent operations (Gaalman

et al 1999) However, this approach is unrealistic as

the complexity of manufacturing processes might be

unavoidable in achieving real-time process plan

gen-eration (Joo et al 2001)

Dong et al (1992) proposed a dynamic

features-based IPPS The product features were extracted using

an AI-based feature extractor with respect to shop

floor conditions Then, rough process plan for a

product was prepared It considers all possible

manufacturing ways for each operations volume

(operation features) that can be produced in one

machine setup considering shop floor capabilities

Moreover, during rough process plan generation,

geometric constraints decide the priority of

manufac-turing for each operation volume Rough process plan

with alternative was input to scheduling Smallest slack

time criterion was considered for scheduling of a batch

size manufacturing shop

Concurrent Manufacturing Planning and shop

control for batch production (COMPLAN), a

Eur-opean ESPRIT project 6805 during the period 1992–

1995, integrates process planning and workshop

scheduling using MPP The COMPLAN approach

was an extension of FlexPlan (Kruth and Detand

1992) The goal of COMPLAN project was to develop

a software system prototype that was capable of

carrying out manual and automatic process planning

and scheduling based on MPP, in a small batch

manufacturing of complex products in a job shop It

contained PPM and workshop scheduling system

(WSS) PPM was capable of handling MPP, and it

could use projected resources load while developing

MPP This module described feasible manufacturing

alternative that provided flexibility to workshop

scheduling WSS followed a hierarchical approach

Usher and Fernandes (1996) proposed a process

planning architecture for integration with scheduling

system It used feature-based approach to planning

and has two phases namely ‘Static Planning’ and

‘Dynamic Planning’ ‘Static Planning’ phase involved

selection, assignment and sequencing of processes and

machines that exist within the shop The output of

‘Static Planning’ phase was a set of alternative

macro-level plans ‘Dynamic Planning’ phase considered

availability of shop floor resources and objectives

specified by scheduler The proposed system was able

to perform for both prismatic and rotational parts.Authors concluded that the proposed two-phasedapproach was able to reduce work load when real-time portion of planning activities were carried out insecond phase

Cho et al (1998) proposed a prototype BlockAssembly Process Planning and Scheduling system inshipbuilding, which consists of a block assembly PPM,

a SM, a bottleneck block selection module and aprocess-replanning module Rule-based reasoningtechnology was applied to determine optimal assemblyunits and assembly sequences in generating initialprocess plans For SM, a schedule revision heuristicwas developed for efficient reallocation of blocks toalternative assembly shops For bottlenecks, blockselection that plays a central role in bridging processplanning and scheduling, a heuristic was developed byemploying an entropy-based partitioning method toidentified bottleneck periods Thus, initial process planwas repeatedly modified and improved by iterating

‘scheduling’ – ‘bottleneck block selection’ – ‘processreplanning’ cycle until workloads were sufficientlybalanced

Sugimura et al (2001) proposed an IPPS systemapplicable to HMS The process planning systemselected suitable sequence of machining feature using

GA approach The optimum sequence of machiningequipment was selected using dynamic programming(DP) after taking into consideration the future schedule

of machining equipments, with objectives to minimisetotal machining and set-up time Machining scheduleswere determined using a real-time scheduling procedure

in which individual job selected suitable machiningequipment, in order to carry out next operation based

on process plan Sugimura et al (2003) extended theirearlier work (Sugimura et al 2001) Here, optimumsequence of machining equipment, two objectivefunctions viz., minimisation of shop time and machin-ing cost were used in DP approach Shop time wascombination of machining time, fixturing time, toolchanging time, transportation time and waiting time.The machining time was calculated on the basis ofmanufacturing process time and operation cost per unittime of manufacturing resources Authors concludedthat the proposed approach was capable to select bothsuitable sequence of machining equipment and machin-ing schedule concurrently

Shrestha et al (2008) developed an IPPS system forHMS using DP method-based modification of processplans Two types of holons viz., job holons andscheduling holons were considered for process plan-ning and scheduling, respectively Based on feasibleprocess plans of all jobs, the scheduling holongenerates the schedule of all equipments Makespan,total machining cost and weighted tardiness cost wereFigure 2 CLA (Zhang and Merchant 1993)

International Journal of Computer Integrated Manufacturing 523

assumed as an objective function for scheduling A

GA-based method was adopted for selecting a

combination of process plans Schedules were

gener-ated using a set of dispatching rules viz., SPT, SPT/

Total Work Remaining and Apparent Tardiness Cost

(ATC) Feedback processes were considered in order

to modify the process plans based on load balancing of

machining equipments Two approaches viz.,

centra-lised approach and DA were developed in order to

modify process plans In centralised approach, the

feedback information from scheduling holon after

scheduling was transferred to job holons and one

modified set of process plans obtained with

considera-tion of constraints of machining equipments In DA,

the job holons modify their process plans without any

centralised control of scheduling holons Results were

compared with and without modification approaches

Authors concluded that in centralised approach of

process plan modification, the makespan was

im-proved for the case where there was concentration of

the machining load on some machining equipments

Also, in the DA, the makespan and weighted tardiness

cost were improved for both cases where there is and

there was no concentration of machining loads on the

machining equipment

Zhang et al (2003a) proposed an IPPS scheme for

batch manufacturing of prismatic parts This approach

is similar to COMPLAN except, the process planning

system could generate whole solution space based on

operations for a given part using SA and GA in order

to find optimal plan (Kempenaers et al 1996) An

‘intelligent facilitator’ was used to generate

instruc-tions for process plan modificainstruc-tions The integration

was achieved through an ‘intelligent facilitator’ that

provide feedback to PPM of a particular job They

developed two algorithms viz., ‘machine utilisation’

and ‘tardy job’ based on heuristic and concluded that

algorithms based on heuristic were suitable to achieve

a satisfactory schedule Wang et al (2008) have

extended the work of Zhang et al (2003a) They

proposed two heuristic algorithms namely fine-tuning

(Tardy) and quick-tuning (QH-Tardy) In

FH-tardy algorithm, solution space of selected operation of

selected tardy job was modified in each iteration In

QH-Tardy algorithm, solution space of selected

operation was modified in each iteration for each

tardy job They concluded that the proposed heuristic

was able to explore process plan solution space in

order to reduce job tardiness

Wong et al (2006a) proposed an agents-based

multi-stage negotiation protocol scheme for IPPS in a

job shop environment The proposed system

com-prised of part agents and machine agents to represent

parts and machines, respectively Part and machine

agents negotiate to establish schedule using process

plans and operation details from AND/OR graph Thenegotiation protocol was established to handle multi-ple task and many-to-many negotiation A currencyconversion function, which incorporates processingtime and due date, was adopted for bidding A java-based simulation model multi-agents negotiation(MAN) was used to implement the proposed ap-proach Authors concluded that in the pursuit of localobjectives such as parts flow time, the proposedapproach performs better than meta-heuristics Wong

et al (2006b) extended their earlier work (Wong et al.2006a) and proposed a hybrid-based multi-agentsystem called Online Hybrid Agents-based Negotiation(oHAN) It comprised of local agents (part andmachine agents) and a supervisor agent The super-visor agent was used for global rescheduling processand influenced the decisions made by local agents Itacted as a system coordinator manager in betweenlocal agents for global rescheduling objective Authorsconcluded that hybrid approach is effective for largerscale rescheduling and provided a better globalperformance They suggested a mobile agent inoHAN in order to handle job order details

2.3 Distributed approach

It performs both process planning and productionscheduling simultaneously It divides process planningand production scheduling tasks into two phases Thefirst phase is preplanning In this phase, processplanning function analyses the job based on theproduct data The features and feature relationshipsare recognised, and corresponding manufacturingprocesses are determined The required machinecapabilities are also estimated The second phase isthe final planning, which matches required job opera-tions with the operation capabilities of availableproduction resource The integration occurs at thepoint when resources are available and the job wasrequired The result is dynamic process planning andproduction scheduling constrained by real-time events.This approach is also referred to as just-in-timeapproach or phased or progressive approach Figure 3shows DA

This approach is the only one that integrates thetechnical and capacity-related planning tasks into adynamic fabrication planning system (Larsen andAlting 1990) However, this approach requires highcapacity and capability from both hardware andsoftware Moreover, scope of DA is limited withinsome specific CAPP functions such as process andmachine selection as detailed process planning tasksare shifted down to manufacturing stages for enhan-cing flexibility (Joo et al 2001) From implementationviewpoint, both process planning and scheduling

departments in a company have to be dismantled and

reorganised (Haddadzade et al 2009)

Aanen et al (1989) proposed a hierarchical

approach for IPPS in a FMS The components of

FMS were integrated with software called supervisory

control system The primary objective was to satisfy

due dates of the order and secondary was to minimise

change over and idle times of machines within time

horizon Initially, planning function was solved and

resulting output becomes the input for scheduling

Feedback information was provided to planning level,

if output of scheduling was not satisfied Authors

planned and scheduled two types of activities viz.,

machining activities and operator activities The time

horizon (of about 10 days) at the planning level was

divided in periods of 1 day For each day, machining

activities to be performed were assigned The resulting

list was called a day list List of activities for 1st day

was the input to scheduling level Scheduling criterion

was to minimise makespan of the day list The

scheduling was performed in two steps: first,

schedul-ing of machinschedul-ing activities (usschedul-ing branch-and-bound

method) and second, scheduling of operators activities

Zhang and Merchant (1993) proposed an integrated

process planning model The modules were integrated

in three levels viz., pre-planning module at initial

integration level, pairing planning module at

decision-making level and final-planning module at functional

integration level Pre-planning module performed three

activities viz., feature reasoning, process recognition

and setup determination The output of pre-planning

was possible setups, machining operation and associate

times Pairing planning module worked in three steps:machining selection, tooling and fixture selection andexact time selection Final planning level worked inthree steps viz., operational tolerance analysis, opera-tion sequencing and overall time and cost calculations

In pre-planning level, SM provides available equipment

in next time window In decision-making level, theavailable equipments were matched with requirement

of setup and matching processes Decision-makingmodule (DMM) was central element to perform bymeans of real-time information Real-time machinedatabase constructed base in the task and time assign-ment of the machine that contained information aboutall machines in shop floor

Huang et al (1995) proposed a progressiveapproach containing three phases namely pre-plan-ning, pairing planning and final planning Theactivities within each phase takes place in differenttime periods Pre-planning was executed in early stage,

as soon as product design finished Pairing planningexecuted in later stage, when an order has beenreleased to the shop and final planning was executedjust before the manufacturing begins The interactionbetween process planning and scheduling takes place inall three phases The model consisted of PPM and SM.PPM was responsible for generating process plansaccording to part design specifications The criterionfor process plan selection was manufacturing leadtime SM was responsible for allocating resources inthe shop and overall management of flow of produc-tion orders They solved IPPS problem by developingfirst mathematical models and then using optimisationFigure 3 DA (Zhang and Merchant 1993)

International Journal of Computer Integrated Manufacturing 525

algorithms They found that computational complexity

was reduced by using progressive approach and to

implement progressive approach, database

manage-ment, machine cell monitoring and knowledge

proces-sing must be well addressed

Kempenaers et al (1996) proposed a collaborative

process planning and scheduling system, which

con-sists of workshop evaluator (WE), schedule evaluator

(SE) and process plan evaluator (PPE) It was based on

(i) the aim of ‘Evaluation Module’ (EM) to improve

quality of delivery work and (ii) the idea of ‘feedback’

at all levels Feedback information was used by

evaluation modules to improve the quality of delivered

output WE captured the performance information of

workshop that was total operation time, machine

breakdown and problems with process plans etc

Workshop disruptions were recorded and appropriate

measures were taken SE evaluates performance of

schedule versus results of workshop A request can be

launched to add an alternative to specific process plan

to bypass a resource that has broken down PPE takes

feedback data into account, and it deals with

produc-tion constraints coming from SE The feedback loop

from scheduling to process planning was based on the

established production constraints A set of constraints

express the demand from scheduling concerning

quality of operation routings ‘General Constraints’

considered current and predicted loading of shop,

while in ‘Specific Constraints’, the scheduler asks

process planning department to regenerate process

plan A constraint-based process planning kernel was

used to integrate the concept of constraint feedback,

that support manual, semi-automatic and automatic

process planning

Mamalis et al (1996) proposed an on-line IPPS

that consists of two phases First phase was an offline

process planning and schedule generation Here,

information flow exists from and into the CAPP and

scheduling system It provides dynamic feedback to

process planning system On the basis of feedback,

MPP were generated by process plan generator The

second decision phase was an on-line process planning

and scheduling in order to consider the disruption at

shop floor A DMM reacts to modifications of the

initial production conditions and provides optimal

scheduling decisions, on the basis of variable routing

and dispatching strategies such as come,

first-serve, SPT and longest processing time (LPT) The

objective was to minimise operation time during

process planning and minimisation of total delay of

parts during scheduling They designed an information

model to maintain and develop an interaction between

process planning and SM The proposed approach was

validated using simulation in job-shop scheduling

environment for machining of rotational parts They

concluded that on-line IPPS considers the dynamicbehaviour of manufacturing system

Gu et al (1997) proposed bidding based MASapproach that has four types of agents namely: part,shop manager, machine and tool The machine agentsperforms process planning, that includes, STEP par-sing and interpretation (to produce data for processplanning), tolerance analysis, operations and setupplanning, machines, tools and fixtures selection,whereas other agents performs the standard functionsrelated with the resources to which they represents.The scheduling was based on the cost model It makesdecisions during the process of negotiations, withconsideration of machining time, setup time, toolchange time, cost of tooling and due date of part Theyconcluded that bidding-based approach was effectiveand efficient for IPPS

Sadeh et al (1998) proposed an Integrated ProcessPlanning/Production Scheduling (IP3S) shell for agilemanufacturing, based on a blackboard architecturethat supports concurrent development and dynamicrevision of integrated process planning schedulingsolutions The system consists of a blackboard, acontroller, a collection of knowledge sources—includ-ing a process planning knowledge sources, a productionscheduling knowledge sources, a communicationknowledge sources and several analysis knowledgesources (e.g a knowledge source to generate resourceutilisation statistics and to evaluate resource contention

in different situations)—and a Motif-based graphicaluser interface (GUI) The solution in IP3S progressedthrough cycles during which one or more unresolvedissues instances were selected to be resolved, aparticular method of resolution selected from amongset of methods applicable to the instance(s), and themethod executed by invoking appropriate knowledgesources

Gindy et al (1999) proposed an IPPS approachbased on concurrent engineering (CE) It contains aknowledge-base facility modelling functions, a feature-based process planning system and a simulation-based

SM Resource element (RE) concept was used forrepresentation of manufacturing environment It de-scribed process capability, which contained informa-tion of commonality and uniqueness of machinesinside factory of form generating schema (FGS).FGS was a combination of cutting tool of specificgeometry, a set of relative motions between a part andcutting tool and technological output During processplanning, a number of alternatives using technologicalsolutions for feature were produced for each feature of

a part Machine tool of manufacturing system wasdescribed by a set of REs A machine-independentGPP was input for scheduling system in order tooptimise machine utilisation and tardiness Final

process plan and production schedule were generated

simultaneously just before the part was about to be

manufactured They compared performance of

ma-chine-based (conventional) system and RE-based

proposed system on the basis of mean tardiness and

average flow time They concluded that proposed

approach was much less sensitive to dispatching rules

and due date assignment approaches Average machine

utilisation of RE-based system (40%) was higher than

machine-based system (22%) RE-based system was

more capable to cope with changes on demand pattern

and machines breakdown

Morad and Zalzala (1999) proposed an approach

using concept of CE with GA in order to

simulta-neously optimise processing capabilities of machines

including processing costs as well as number of rejects

in alternative machines with scheduling of jobs The

chromosome consists of order of parts as well as order

of operations and associated machines to perform

operations The formulation was based on

multi-objective weighted sums optimisation The multi-objectives

were to minimise total rejects produced, to minimise

total cost of production and to minimise makespan

and weight for total rejects produced, total cost of

production and makespan were taken as w1, w2 and

w3, respectively The scheduling was based on SPT

dispatching rule Authors compared their weighted

sum approach with traditional and multi-objective GA

(MOGA) approach They concluded that weighted

sum approach produced better results indicating an

improvement of makespan and total number of rejects

produced compared with conventional method

Chan et al (2001) proposed MAS-based

frame-work for integrated, distributed and co-operative

process planning system called IDCPPS The tasks

were separated into three levels viz., initial planning

level, decision-making level and detail planning level

Initial planning level involved: features reorganisation,

selection of machining processes, generation of process

sequences and manufacturability evaluation The

result of this level was a set of alternative process

plans Decision-making level was interacted with

scheduling system in order to consider availability of

shop floor resources The result of this level was a set

of ranked near-optimal alternative process plans

Detailed planning level included: tool selection, final

machine selection, machine parameters determination

and calculation of estimates for both machining cost

and time The output of this level was the final detailed

linear process plans The integration with scheduling

was considered at each stage with process planning

Authors concluded that, firstly, IDCPP system was a

tool for CAD by integrating down-stream constraints

into the design phase Secondly, it integrated process

planning and scheduling Thirdly, it responded

continually to manufacturing conditions and tion tasks change with the help of MAS

produc-Wu et al (2002) proposed a framework for IPPS in

a digital virtual manufacturing (DVM) environmentusing Java-based MAS A cost function was proposedfor optimal partner selection in virtual enterprise,which considered partner’s manufacturing capability,processing time, partners location and part due date.The framework was divided into two phases viz.,enterprise phase and partner phase Enterprise phaseselected optimal manufacturing partner in a DVM.The process plan, at this level, consists of set-up inorder to identify the machining process for features Atthis level, scheduling provided the information onprocess potential of particular partner through capa-city planning The second phase performed shop floor-level integration At this level, PPM checked themanufacturing cell capacity maintained by SM fromtime to time, compared cell level costs and selectedsuitable machine cells for part manufacturing Authorsconcluded that cost function found to be effective foroptimal partner selection in a DVM

Zhang et al (2003b) proposed a concurrentintegrated process planning system (CIPPS), to inte-grate CAD, process planning and scheduling throughworking of process planning and scheduling system inparallel at three levels viz., initial planning level,decision making level and detailed planning level.CIPPS was implemented in a HMS consortium (ahierarchical architecture of self-consistent, cooperativemodules) in order to improve flexibility, expandabilityand stability They concluded that CIPPS suits therequirements of agile manufacturing such as response

to changing conditions of organisation; reduction intime and cost of processes and integration withdifferent domains of manufacturing

Wang et al (2003) proposed an approach called

‘distributed process planning’ (DPP), based on designfor machining concept They used machining feature-based reasoning, MAS-based decision making andfunction block-based computer numerical control.DPP architecture was designed based on two-layerstructure for supervisory planning and operationplanning It had three major sub-systems for design,planning and control, which shared one dynamicdatabase A real-time network and secured factorynetwork enabled the functionality of each module.Authors concluded that IPPS can be done early atsupervisory planning level where machine selection wasconducted The dynamism of process planning enabled

by DPP methodology is crucial to next-generationreconfigurable manufacturing system (RMS), and it isbeneficial to apply DPP to the RMS framework.Wang and Shen (2003) discussed the agent-baseddecision making, machining feature-based processInternational Journal of Computer Integrated Manufacturing 527

planning and function block-based process execution

associate with DPP An example of pocket milling in

order to generate a function block-based process plan

was presented Moreover, Wang et al (2005) proposed

a framework of collaborative process planning

sup-ported by a real-time monitoring system Authors

concluded that an adaptive process plan of the part can

be generated by converting its machining features to

appropriate function blocks designs Details on

func-tion block design in DPP can be found in Wang et al

(2006) Wang et al (2009) presented a function block

technology-based adaptive approach for design and

integration of event-driven function blocks with

process/set-up planning and execution monitoring

Wang (2009) developed an integrated system for

web-based collaborative planning and control, supported by

real-time monitoring for dynamic scheduling

Cai et al (2009) proposed to solve IPPS problem

indirectly through a multi-machine setup planning

approach using GA, which utilised the adaptability of

process plan associated with setups A tool accessibility

examination approach was used for adaptive setup

planning (ASP), and it was extended to solve

multi-machines setups planning problem The proposed

concept was validated with a part having four three

axis-based setups to be machined on three machines

Authors concluded that GA-based ASP is capable to

quickly respond in changing shop floor situations

Wang et al (2010) proposed an IPPS approach for

job shop machining operations via a two-step ASP

using GAs It consists of generic setup planning (step

one) and adaptive setup merging (step two) in order to

optimise cost, quality, makespan and machines

utilisa-tion Initially, three-axis-based setup plans were

created and then merge some setups in order to create

final setup with consideration of available machines

Final specific process plan was created after scheduling

and setup merging An example part of 26 machining

features and three machines were considered to

validate the proposed algorithm Authors concluded

that proposed approach can generate setup plans

adaptively based on machines availability and

capability

Zattar et al (2008) proposed a hierarchical MAS

using feature-based time-extended negotiations

proto-col for decision making about real-time adaptation of

process plan with alternatives in order to minimise

makespan and mean flow time in a job shop

environment The total operation time as reported in

Kim et al (2003) was divided in three parts: 20%

processing time in the machine, 70% machine setup

and 10% fixturing setup They found that mean

makespan and mean flow time were 426.68 minutes

and 329.59 minutes, respectively, without

considera-tion of setup time, while the values of mean makespan

and mean flow time were 374.46 minutes and 286.96minutes with consideration of setup time Theyconcluded that global average values of makespanand flow time obtained by the proposed approach werebetter than those generated by SEA of Kim et al.(2003) Zattar et al (2010) extended their earlier work(Zattar et al 2008) Total operation time as reported inKim et al (2003) was divided in three parts: 80%processing time in the machine, 10% machine setupand 10% fixturing setup They found that meanimproved rate of makespan and flow time was 3.66%and 9.13%, respectively, without consideration ofsetup time, while the values of mean improved rate

of makespan and flow time was 6.18% and 10.56%with consideration of setup time They concluded thatreduction of both objectives due to reduction innumber of machines changes on which the jobs weremanufactured They also concluded that assumption ofinclusion of set-up time in total processing time ofmachines may result in an incorrect analysis of theproblem

Ueda et al (2007) proposed an evolutionary ANN(EANN), to produce simultaneous decision of processplanning and scheduling The local decisions such asselection of part to process and selection of machinefor selected part were taken by machine agents Eachmachine learns the state of shop floor using EANN.System objective was achieved by multiplications ofresult of each machines learning Li et al (2008a)proposed a cooperative process planning and schedul-ing system (CPPS) for IPPS in a job shop environment.The system equipped with three game theory strategiesviz., Pareto strategy, Nash strategy and Stackelbergstrategy in order to minimise makespan, job tardiness,manufacturing cost and load balancing of machines.Three algorithms were applied to CPPS problem viz.,PSO, SA and GA They considered machine break-down and new order arrival as dynamic features of ajob shop Authors concluded that lower manufacturingcost can be achieved through utilisation of cheapmachines, but it is conflicting with the criterion ofbalanced level of machine utilisation Moreover, SA-based algorithm took shorter time to find goodsolution, but it was dependent on its parameters andthe problem to be optimised GA and PSO-basedalgorithms were slow in finding good solutions, butthey were robust for optimisation problem

Zhanjie and Ju (2008) proposed a GA-based IPPSsystem in which process route was selected on the basis

of balanced level of machines utilisation, minimumprocessing cost and SPT dispatching rules AND/ORnetwork was used to represent MPP They found thatfor 70–80% of test-bed problems, as reported in (Kim

et al 2003) proposed GA provided the best mance among the algorithm compared Shukla et al

(2008) proposed a bidding based MAS in which tool

cost was considered as dynamic quantity rather than a

constant Tool cost comprises tool-using cost and tool

repairing cost Tool cost was predicted by data mining

agent When a job arrives at shop floor, the component

agent announces a bid for one feature at a time to all

machines agents Once all features were assigned to

appropriate machine, this information was utilised by

optimisation agent to find optimal process plan and

schedule using hybrid TS-SA algorithm

Li et al (2009, 2010a) proposed an agent-based

approach with an optimisation agent and a

mathema-tical model for IPPS in a job shop environment The

system contained three agents and databases Job

agents and machine agents were used to optimise

alternative process plan and schedule All three

flexibilities namely operational, sequential and

proces-sing flexibility were used in process planning The

objective of process planning system was to minimise

production time, whereas for scheduling system, it was

to minimise makespan When changes occur at shop

floor and determined schedule cannot be carried out,

machine agents negotiate with other agents (including

job agents and optimisation agents) in order to trigger

a rescheduling process

2.4 Other approaches

There are several contributions in literatures that do

not fall in any of the above category These

contribu-tions are discussed below

Liao et al (1993) proposed the modification of an

existing CAPP system, i.e computer managed process

planning (CMPP) – for achieving CAPP/scheduling

integration The primary functional areas of CMPP

that should be modified for CAPP/scheduling

integra-tion were process decision model file (PDMF) and

machine tool file PDMF contained process decision

rules in order to determine process plans An

opera-tion-machine index was developed for selecting best

machine from all alternates in process planning stage

to satisfy two scheduling criterion, first was to

minimise mean flow time and second was to reduce

number of jobs tardy The modifications were: (i) the

creation of a secondary machine tool file containing

data needed to calculate operation-machine index and

(ii) a software program for modifications of process

decision rules They concluded that total processing

time was reduced by 6% resulting from the use of

altered PDMF Authors concluded that IPPS can be

achieved through modifications of an existing CAPP

system, instead of developing a new system

Zijm and Kals (1995) proposed IPPS approach in a

small batch manufacturing shop A set of initial

process plans and a resource decomposition procedure

were used to determine schedule that minimiselateness If, schedule performed unsatisfactory, acritical path analysis was conducted to select jobs ascandidates for MPPs The critical path graph auto-matically selects an operation, which causes the largestlateness The procedure and algorithms were imple-mented in a multi-resources shop floor planning andscheduling system, called ‘job planner’ The systemcontained three layers viz., an automatic scheduler, aninteractive scheduling mode and monitoring andcontrol system ‘Job-planner’ attempted to minimiselateness by calculating paths for each possible sche-dule The methodology has been tested at a machineshop at Ureno in Almelo (The Netherlands), andexperiments yielded an improvement of manufacturinglead-time of 10–30%

Chan et al (2006) proposed an artificial immunesystem (AIS)-based algorithm inherited with fuzzylogic controller (FLC) to solve IPPS problem Theproposed algorithm could handle multiple ordersinvolving outsourcing strategy They considered man-ufacturing system with alternative operations se-quences, alternative machines for different operationsand precedence relationship between the operations.The system was modelled as TSP with precedencerelationship The feasible operation sequence wasgenerated by merging features of AIS-FLC, directedgraph and topological sort techniques The objectivefunction was to minimise makespan and simulta-neously considering due date of customer orders Theproposed algorithm was tested with five machines withone outsourcing machine The authors concluded thatthe proposed concept of outsourcing strategy isadvantageous, only when the total transportationtime involved during the process is less than thewaiting time of the part The outsourcing strategyeffectively reduces makespan Chan et al (2009)extended their earlier work (Chan et al 2006) andproposed an enhanced swift converging simulatedannealing (ESCSA) algorithm for scheduling Theproposed algorithm was compared with other optimi-sation methods such as GA, SA, TS and TS-SAalgorithms and found that makespan came out 30 and

55 units for due date45 and 75, respectively, whichoutperformed comparatively Authors concluded thatproposed algorithm was superior and a simple plan-ning tool to strategically select outsourcing machineand perform operations on them while consideringtechnological constraints of real shop floor situations.Guo et al (2009) proposed a PSO algorithm andreplanning method for machine breakdown status andnew order arrival The solutions were encoded intoPSO particles to search for best sequence of operationsthrough optimisation strategies of PSO algorithm.Authors concluded that PSO outperformed both GAInternational Journal of Computer Integrated Manufacturing 529

and SA by considering computational efficiency The

proposed algorithm showed improvement in

perfor-mance by applying crossover operator taken from GA

Shao et al (2009) suggested an approach by

synthesis-ing integration methodology of NLA and DA in which

process planning and scheduling system were working

simultaneously A simulation approach based modified

GA was used The objective of process planning was to

minimise production time However, two objective

functions of scheduling were considered First was to

minimise makespan and second was synthesis

con-sideration of makespan and balanced level of machine

utilisation Authors found that proposed approach

was better than hierarchical approach

Li and Ierapetritou (2009) reviewed various

inte-gration methodologies and uncertainties for planning

and process scheduling in the process industries

Authors concluded that integration of planning and

scheduling and systematic considerations of

uncertain-ties have a tremendous impact on industries in order to

increase their production flexibilities

Sormaz et al (2010) proposed a model for

integration of product design with process planning

and scheduling information in real time using

XML-based data representation It involves features

mapping from CAD file, process selection for partdesign and integration with scheduling and simulation

of FMS model utilising alternative routings Twoscenarios viz., integration of CAPP and scheduling andintegration of CAPP with FMS control simulationwere considered in order to demonstrate the proposedmodel on a group of nine prismatic parts Features-based model was created in Unigraphics CADpackage Rule-based system was used for processselection and selection of alternative machine availablewith tool and processing time information Schedulingapplication used LP approach for simultaneous selec-tion of alternatives for each feature and their schedul-ing FMS simulation model was developed as a discreteevent model in Arena A feature focused dynamicmodel was developed to integrate all three modules offeature mapping, process planning and FMS control-ler The simulation model collects the current machineutilisation level and queue size for every machine andsends this data to FMS controller Authors concludedthat proposed approach may shorter developmentcycle in integration of various manufacturing modules.Table 1 summarise the main features of NLA, CLAand DA Table 2 provides list of papers reviewed in thepresent work

Table 1 Features of integration of process planning and scheduling approaches

S No

Integration

1 NLA 1 Process plans contain alternative routing, which offer high degree of flexibility to scheduling

2 It contains possibilities of improving off-line scheduling performance and can be quickly react todisturbances on the shop floor

3 It can be implemented in a company that has process planning and scheduling departments

4 It has one-way of information flow i.e from process planning to production planning Therefore, itmay be impossible to achieve full optimal results in integrating two functions

5 Some of the process plans created are not feasible according to real time shop status

6 Considering all possible process alternatives for resource allocation may enormously increasescomplexity of process plan representation

2 CLA 1 Each generated process plan is feasible and based on current shop floor conditions

2 It enhances real time, intuition and manipulability of process planning system

3 The real-time status of manufacturing system is essential for it

4 It requires high-capacity software and hardware

5 The process planning and scheduling departments of a company may have to dismantle andreorganise to take the full advantage

6 The adaptation of a step-by-step local view limits the solution space for subsequent operations

3 DA 1 It completely integrates process planning and scheduling functions and provides the reasonable

schedules without generating superfluous process plans

2 It performs process planning and scheduling in parallel

3 The activities within each phase take place in different time periods

4 The interaction between process planning and scheduling starts from a more global level and ends at

a more detailed level

5 It requires high-capacity software and hardware

6 Process planning and scheduling departments of a company have to be dismantled and reorganised

7 It has limited scope within some specific CAPP function such as process and machine selection asdetailed process planning tasks are shifted down to manufacturing stages for enhancing flexibility

8 It is truly integrated approach with whole solution space available but, due to vast solution space,finding a feasible solution in a reasonable amount of time is difficult

Note: NLA, non-linear approach; CLA, closed-loop approach; DA, distributed approach.

3 Conclusions

IPPS plays a cardinal role for performance

improve-ment in a manufacturing system In this article, we

have provided a state-of-the-art review on the IPPS

Various approaches to IPPS along with their

advan-tages and disadvanadvan-tages have been discussed

More-over, contributions of different researchers have been

presented briefly We hope that the present work will

provide a platform to new researchers Having

reviewed the literature, we would like to add brief

comments on IPPS

(1) Multi-agents system approach is found to be

the most promising DA, as they are recognised

as an effective way to realise IPPS adaptiveness

Agents-based approach may perform better

when number of agents and level of

negotia-tions are less However, these approaches may

fail to perform when number of agents involved

as well as level of negotiations are large (i.e

system size is large) as they have a limited

effective negotiation mechanism In such

sce-nario, there is need to increase the capability of

supervisor agent so that effective negotiation

among resources agents can be achieved

(2) Most of the researchers considered single

objective to solve IPPS problem Single

objective are not able to represent real facturing environment completely Thus, there

manu-is a need to consider multi-objective such asmakespan, mean flow time, tardiness of partssimultaneously as multi-objective representsreal manufacturing environment better thansingle objective Although few studies haveconsidered multi-objectives but more investiga-tions are needed in this direction The viewpoint is in line with the earlier studies (Chan

et al 2006,2009)

(3) Most of the previous research has not rated shop floor disturbances in the IPPS model.Generally, two types of disturbances occurs onshop floor viz., internal and external Internaldisturbances involves machines breakdown, arri-val of new machine, routine maintenance, toolfailure, etc Whereas, external disturbancesinvolves order cancellations, rush order arrivaland change in demand pattern Occurrence ofinternal and/or external disturbances will prob-ably make the current schedule infeasible Thus,there is a need to develop an adaptive-initiativeand interactive IPPS model that is able tominimise the effects of shop floor disturbances.(4) Most of the previous research utilised anoptimisation algorithm However, as the searchspace is larger, computation time increases

incorpo-Table 2 Papers reviewed as per classifications

Chryssolouris and Chan (1985) Dong et al (1992) Aanen et al (1989) Liao et al (1993) Sundaram and Fu (1988) Kruth and Detand (1992) Zhang and Merchant (1993) Zijm and Kals (1995) Tonshoff et al (1989) Usher and Fernandes (1996) Huang et al (1995) Chan et al (2006, 2009) Srihari and Greene (1990) Cho et al (1998) Kempenaers et al (1996) Guo et al (2009) Jablonski et al (1990) Sugimura et al (2001, 2003) Mamalis et al (1996) Shao et al (2009)

Sormaz et al (2010) Kim and Egbelu (1999) Wang et al (2008) Sadeh et al (1998)

Aldakhilallah and Ramesh (1999) Wong et al (2006a, 2006b) Gindy et al (1999)

Weintraub et al (1999) Shrestha et al (2008) Morad and Zalzala (1999)

2009 and 2010) Cai et al (2009)

Note: NLA, non-linear approach; CLA, closed-loop approach; DA, distributed approach.

International Journal of Computer Integrated Manufacturing 531

dramatically Thus, there is a need to develop

hybrid optimisation approach that can find

optimal solution quickly In this direction, a

combination of heuristics and meta-heuristics

techniques will be helpful

(5) There is a need to develop an IPPS approach

that can be implemented in a company with

existing process planning and scheduling

depart-ments and that can integrate process planning

and scheduling quickly This may be done by

combining various approaches such as

NLA-CLA and NLA-DA, so that the developed

approach is able to take advantage of combining

approaches This may result in an improved level

of information exchange between process

plan-ning and scheduling departments

(6) Previous research has not focussed on the

development of industrial applicable IPPS

system In real manufacturing environment,

there exits several practical issues such as job

releasing, machine and tool changes Tool

failures constraints and subassembly levels of

each job, set up time, transportation time, finite

buffers and other resources constraints

Acknowledgement

This research is supported by Government of India (GOI),

Science and Engineering Research Council (SERC),

Depart-ment of Science and Technology (DST), New Delhi, India (SR/

SR3/MERC-098/2007) We would like to thank the anonymous

referees, reviewers and editor for their valuable comments and

recommendations for improving the content of this paper

References

Aanen, E., Gaalman, G.J., and Nawijn, W.M., 1989

Planning and scheduling in an FMS Engineering Costs

and Production Economics, 17 (1–4), 89–97

Aldakhilallah, K.A and Ramesh, R., 1999

Computer-integrated process planning and scheduling (CIPPS):

intelligent support for product design, process planning

and control International Journal of Production

Re-search, 37 (3), 481–500

Baykasoglu, A and O¨zbakır, L., 2009 A grammatical

optimization approach for integrated process planning

and scheduling Journal of Intelligent Manufacturing, 20

(2), 211–221

Benjaafar, S and Ramakrishnan, R., 1996 Modeling,

measurement and evaluation of sequencing flexibility in

manufacturing systems International Journal of

Produc-tion Research, 4 (5), 1195–1220

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

setup planning toward process planning and scheduling

integration International Journal of Production Research,

47 (10), 2745–2766

Chan, F., Kumar, V., and Tiwari, M., 2006 Optimizing the

performance of an integrated process planning and

scheduling problem: an AIS-FLC based approach’ In:

Proceedings of 2006 IEEE Conference on cybernetics and

intelligent systems, 7–9 June, Bangkok, Thailand, 1–8

Chan, F., Kumar, V., and Tiwari, M., 2009 The relevance ofoutsourcing and leagile strategies in performance opti-mization of an integrated process planning and schedul-ing model International Journal of Production Research,

47 (1), 119–142

Chan, F., Zhang, J., and Li, P., 2001 Modelling of grated, distributed and cooperative process planningsystem using an agent-based approach Proceedings of theInstitution of Mechanical Engineering, Part B: Journal ofEngineering Manufacturing, 215 (10), 1437–1451.Cho, K.K., et al., 1998 An integrated process planning andscheduling system for block assembly in shipbuilding.Annals of the CIRP, 47 (1), 419–422

inte-Choi, H and Park, B., 2006 Integration of process planningand job shop scheduling using genetic algorithm In:Proceedings of 6th WSEAS international conference onsimulation, modelling and optimization, 22–24 September,Lisbon, Portugal, 13–18

Chryssolouris, G and Chan, S., 1985 An integratedapproach to process planning and scheduling Annals ofthe CIRP, 34 (1), 413–417

Dong, J., Jo, H.H., and Parsaei, H.R., 1992 A feature-baseddynamic process planning and scheduling Computers &Industrial Engineering, 23, 141–144

Gaalman, G.J.C., Slomp, J., and Suresh, N.C., 1999.Towards an integration of process planning and produc-tion planning and control for flexible manufacturingsystems International Journal of FMS, 11, 5–17.Gindy, N., Saad, S., and Yue, Y., 1999 Manufacturingresponsiveness through integrated process planning andscheduling International Journal of Production Research,

37 (11), 2399–2418

Gliffer, B and Thompson, G.L., 1960 Algorithms forsolving production scheduling problems InternationalJournal of Operational Research, 8, 161–171

Grabowik, C., Kalinowski, K., and Monica, Z., 2005.Integration of the CAD/CAPP/PPC systems Journal

of Materials Processing Technology, 164–165 (2),1358–1368

Gu, P., Balasubramanian, S., and Norrie, D., 1997 based process planning and scheduling in a multi-agentsystem Computers & Industrial Engineering, 32 (2),477–496

Bidding-Guo, Y.W., et al., 2009 Optimisation of integrated processplanning and scheduling using a particle swarm optimi-sation approach International Journal of ProductionResearch, 47 (14), 3775–3796

Haddadzade, M., Razfar, M.R., and Farahnakian, M., 2009.Integrating process planning and scheduling for pris-matic parts regard to due date In: Proceedings of worldacademy of science, engineering and technology, 23–25March, Hong Kong, China, No.51, 64–66

Huang, S.S., Zhang, H.C., and Smith, M.L., 1995 A gressive approach for the integration of process planningand scheduling IIE Transactions, 27, 456–464

pro-Iwata, K and Fukuda, Y., 1989 A new proposal of dynamicprocess planning in machine shop In: Proceedings ofCIRP international workshop on computer aided processplanning, 21–22 September 1989, Hanover University,Germany, 73–83

Jablonski, S., Reinwald, B., and Ruf, T., 1990 Integration ofprocess planning and job shop scheduling for dynamicand adaptive manufacturing control In: Proceedings ofRensselaer’s 2nd international conference on computerintegrated manufacturing, 21–23 May, Troy, NY, USA,444–450

Jain, A., Jain, P., and Singh, I., 2006 An integrated scheme

for process planning and scheduling in FMS

Interna-tional Journal of Advanced Manufacturing Technology,

30, 1111–1118

Joo, J., Park, S., and Cho, H., 2001 Adaptive and dynamic

process planning using neural networks International

Journal of Production Research, 39 (13), 2923–2946

Kempenaers, J., Pinte, J., and Detand, J., 1996 A

collaborative process planning and scheduling system

Advances in Engineering Software, 25, 3–8

Khoshnevis, B and Chen, Q., 1990 Integration of process

planning and scheduling functions Journal of Intelligent

Manufacturing, 1, 165–176

Kim, K and Egbelu, P., 1999 Scheduling in a production

environment with multiple process plans per job

Interna-tional Journal of Production Research, 37 (12), 2725–2753

Kim, Y., Park, K., and Ko, J., 2003 A symbiotic

evolu-tionary algorithm for the integration of process planning

and job shop scheduling Computers and Operations

Research, 30, 1151–1171

Kruth, J.P and Detand, J., 1992 A CAPP system for

nonlinear process plans CIRP Annals - Manufacturing

Technology, 41 (1), 489–492

Kumar, M and Rajotia, S., 2003 Integration of scheduling

with computer aided process planning Journal of

Materials Processing Technology, 138, 297–300

Kumar, M and Rajotia, S., 2006 Integration of process

planning and scheduling in a job shop environment

International Journal of Advanced Manufacturing

Tech-nology, 28 (1–2), 109–116

Larsen, N.E and Alting, L., 1990 Simulations engineering

within process and production planning In: Pacific

Conference on Manufacturing, 17–21 December 1990

Australia

Larsen, N.E and Alting, L., 1992 Dynamic planning

enriches concurrent process and production planning

International Journal of Production Research, 30 (8),

1861–1876

Lee, H and Kim, S., 2001 Integration of process planning

and scheduling using simulation based genetic

algo-rithms International Journal of Advanced Manufacturing

Technology, 18, 586–590

Leung, C.W., et al., 2010 Integrated process planning and

scheduling by an agent-based ant colony optimization

Computers & Industrial Engineering, 59 (1), 166–180

Li, W and McMahon, C., 2007 A simulated

annealing-based optimization approach for integrated process

planning and scheduling International Journal of

Com-puter Integrated Manufacturing, 20 (1), 80–95

Li, W., et al., 2008a Game theory-based cooperation of

process planning and scheduling CSCWD2008, In: 12th

international CSCWD, 16–18 April, Xi’an, China, 841–

845

Li, X., et al., 2010b A review on integrated process planning

and scheduling International Journal Manufacturing

Research, 5 (2), 161–180

Li, X., et al., 2008b A genetic algorithm for integration of

process planning and scheduling problem Lecture Notes

in Artificial Intelligence, 5315, 495–502

Li, X., et al., 2009 Multi-agent based integration of process

planning and scheduling In: 13th international conference

on computer supported cooperative work in design, 22–24

April, Santiago, Chile, 215–220

Li, X., et al., 2010c An effective hybrid algorithm for

integrated process planning and scheduling International

Journal of Production Economics, 126 (2), 289–298

Li, X., et al., 2010a An agent-based approach for integratedprocess planning and scheduling Expert Systems WithApplications, 37, 1256–1264

Li, Z and Ierapetritoua, M., 2009 Integration of Planningand Scheduling and Consideration of Uncertainty inProcess Operations, Computer Aided Chemical Engineer-ing In: 10th international symposium on process systemsengineering: Part A, 16–20 August, Salvador-Bahia,Brazil 27, 87–94

Liao, T.W., et al., 1993 Modification of CAPP systems forCAPP/scheduling integration Computers & IndustrialEngineering, 25 (1–4), 203–206

Mamalis, A.G., Malagardis, I., and Kambouris, K., 1996.On-line integration of a process planning module withproduction scheduling International Journal of AdvanceManufacturing Technology, 12, 330–338

Moon, C., Kim, J., and Hur, S., 2002 Integrated processplanning and scheduling with minimizing total tardiness

in multi-plants supply chain Computers and IndustrialEngineering, 43, 331–349

Moon, C., et al., 2008 Integrated process planning andscheduling in a supply chain Computers & IndustrialEngineering, 54 (4), 1048–1061

Morad, N and Zalzala, A., 1999 Genetic algorithms inintegrated process planning and scheduling Journal ofIntelligent Manufacturing, 10, 169–179

Palmer, G., 1996 A simulated annealing approach tointegrated production scheduling Journal of IntelligentManufacturing, 7, 163–176

Potter, M.A., 1997 The design and analysis of a tional model of cooperative coevolution Thesis (PhD).George Mason University, USA

computa-Rajkumar, M., et al., 2010 A GRASP algorithm for theintegration of process planning and scheduling in aflexible job-shop International Journal of ManufacturingResearch, 5 (2), 230–251

Sadeh, N., et al., 1998 A blackboard architecture for integratingprocess planning and production scheduling ConcurrentEngineering: Research and Applications, 6 (2), 88–100.Saygin, C and Kilic, S., 1999 Integrating flexible processplans with scheduling in flexible manufacturing systems.International Journal of Advanced Manufacturing Tech-nology, 15, 268–280

Shao, X., et al., 2009 Integration of process planning andscheduling: a modified genetic algorithm-based ap-proach Computers and Operations Research, 36, 2082–2096

Shen, W., Wang, L., and Hao, Q., 2006 Agent-baseddistributed manufacturing process planning and schedul-ing: a state-of-the-art survey IEEE Transactions onSystems, Man, and Cybernetics–Part C: Applicationsand Reviews, 36 (4), 563–577

Shrestha, R., et al., 2008 A study on integration of processplanning and scheduling system for holonic manufactur-ing with modification of process plans InternationalJournal of Manufacturing Technology and Management,

14 (3–4), 359–378

Shukla, S., Tiwari, M., and Son, Y., 2008 Bidding-basedmulti-agent system for integrated process planning andscheduling: a data-mining and hybrid tabu-SA algo-rithm-oriented approach International Journal of Ad-vanced Manufacturing Technology, 38, 163–175

Sormaz, D.N., et al., 2010 Integration of product design,process planning, scheduling, and FMS control usingXML data representation Robotics and Computer-Inte-grated Manufacturing, DOI:10.1016/j.rcim.2010.07.014.International Journal of Computer Integrated Manufacturing 533

Srihari, K and Greene, T.J., 1990 MACRO-CAPP: a

prototype CAPP system for an FMS International

Journal of Advanced Manufacturing Technology, 5, 34–51

Sugimura, N., Hino, R., and Moriwaki, T., 2001 Integrated

process planning and scheduling in holonic

manufactur-ing systems In: Proceedmanufactur-ings of IEEE international

symposium on assembly and task planning soft research,

28–29 May, Fukuoka, Japan, 4, 250–254

Sugimura, N., Shrestha, R., and Inoue, J., 2003 Integrated

process planning and scheduling in holonic

manufactur-ing systems - optimization based on shop time and

machining cost In: Proceeding of the 5th IEEE

Interna-tional Symposium on Assembly and Task Planning, 10–11

July, Besanqon, France, 36–41

Sundaram, R.M and Fu, S.S., 1988 Process planning and

scheduling Computer and Industrial Engineering, 15,

296–307

Tan, W and Khoshnevis, B., 2000 Integration of process

planning and scheduling – a review Journal of Intelligent

Manufacturing, 11, 51–63

Tonshoff, H.K., Beckendorff, U., and Andres, N., 1989

FLEXPLAN: a concept for intelligent process planning

and scheduling In: CIRP International Workshop, 21–22,

September, Hannover, Germany, 87–106

Ueda, K., Fuji, N., and Inoue, R., 2007 An emergent

synthesis approach to simultaneous process planning and

scheduling Annals of CIRP, 56 (1), 463–466

Usher, J and Fernandes, K., 1996 Dynamic process

planning – the static phase Journal of Materials

Processing Technology, 61, 53–58

Usher, J., 2003 Evaluating the impact of alternative plans on

manufacturing performance Computers and Industrial

Engineering, 45, 585–596

Wang, J., et al., 2009 Reducing tardy jobs by integrating

process planning and scheduling functions International

Journal of Production Research, 47 (21), 6069–6084

Wang, L and Shen, W., 2003 DPP: an agent-based

approach for distributed process planning Journal of

Intelligent Manufacturing, 14 (5), 429–439

Wang, L., 2009 Web-based decision making for

collabora-tive manufacturing International Journal of Computer

Integrated Manufacturing, 22 (4), 334–344

Wang, L., Cai, N., and Feng, H., 2009 Function blocks

enabled dynamic set-up dispatching and execution

monitoring International Journal of Computer Integrated

Manufacturing, 22 (1), 3–12

Wang, L., et al., 2010 ASP: an adaptive setup planning

approach for dynamic machine assignments IEEE

Transactions on Automation Science and Engineering, 7

(1), 2–14

Wang, L., Feng, H.Y., and Cai, N., 2003 Architecture design

for distributed process planning Journal of

Manufactur-ing Systems, 22 (2), 99–115

Wang, L., Jin, W., and Feng, H.Y., 2006 Embedding

machining features in function blocks for distributed

process planning International Journal of Computer

Integrated Manufacturing, 19 (5), 443–452

Wang, L., Song, Y., and Shen, W., 2005 Development of a

function block designer for collaborative process

plan-ning In: Proceedings of the ninth international conference

on computer supported cooperative work in design, 24–26

May, Coventry, UK, 1, 217–222

Wang, Y.F., et al., 2008 An integrated approach to reactive

scheduling subject to machine breakdown In: Proceeding

of IEEE International Conference on Automation and

Logistics, 1–3 September Qingdao, China, 542–547

Weintraub, A., et al., 1999 Scheduling with alternatives: alink between process planning and scheduling IIETransactions, 31, 1093–1102

Wong, T., et al., 2006b Integrated process planning andscheduling/rescheduling an agent-based approach Inter-national Journal of Production Research, 44 (18–19),3627–3655

Wong, T.N., et al., 2006a An agent-based negotiationapproach to integrate process planning and scheduling.International Journal of Production Research, 44 (7),1331–1351

Wu, S., Fuh, J., and Nee, A., 2002 Concurrent processplanning and scheduling in distributed virtual manufac-turing IIE Transactions, 34, 77–89

Yang, Y., Parsaei, H., and Leep, H., 2001 A prototype of afeature-based multiple-alternative process planning sys-tem with scheduling verification Computers and Indus-trial Engineering, 39, 109–124

Zattar, I., et al., 2008 Integration between process planningand scheduling using feature-based time-extended nego-tiation protocols in a multi agent system InternationalJournal of Services Operations and Informatics, 3 (1), 71–89

Zattar, I., et al., 2010 A multi-agent system for the gration of process planning and scheduling usingoperation-based time-extended negotiation protocols.International Journal of Computer Integrated Manufac-turing, 23 (5), 441–452

inte-Zhang, H.C and Merchant, M.E., 1993 IPPM – a prototype

to integrate process planning and job shop schedulingfunctions CIRP Annals - Manufacturing Technology, 42(1), 513–518

Zhang, J., et al., 2003b A holonic architecture of the current integrated process planning system Journal ofMaterials Processing Technology, 139, 267–272

con-Zhang, Y., Saravanan, A., and Fuh, J., 2003a Integration ofprocess planning and scheduling by exploring theflexibility of process planning International Journal ofProduction Research, 41 (3), 611–628

Zhanjie, W and Ju, T., 2008 The research about integration

of process planning and production scheduling based ongenetic algorithm CSSE, International conference oncomputer science and software engineering, 1, 9–12.Zhao, F., et al., 2004 A genetic algorithm based approachfor integration of process planning and productionscheduling In: Proceedings of international conference

on intelligent mechatronics and automation, 26–31 August,Chengdu, China, 483–488

Zhao, F., et al., 2010 A hybrid particle swarm optimisationalgorithm and fuzzy logic for process planning andproduction scheduling integration in holonic manufac-turing systems International Journal of Computer Inte-grated Manufacturing, 23 (1), 20–39

Zhao, F., et al., 2006 Integration of process planning andproduction scheduling based on a hybrid PSO and SAalgorithm In: Proceedings of IEEE international con-ference on mechanical and automation, 25–28 June,Luoyang, China, 2290–2295

Zijm, W.H.M and Kals, H.I.J., 1995 The integration ofprocess planning and shop floor scheduling in smallbatch part manufacturing CIRP Annals - ManufacturingTechnology, 44 (1), 429–432

A method for engineering design change analysis using system modelling and knowledge

management techniquesGenyuan Feia, James Gaoa*, Oladele Owodunniaand Xiaoqing Tangb

Keywords: engineering design; change management; system modelling; design conflict solving; knowledgemanagement system

1 Introduction

Engineering changes have been recognised as

inevita-ble in complex engineering product development

(Huang et al 2003, Palani Rajan et al 2005, Keller

et al 2009) They have great influences on downstream

developing and production activities, thus making

product development very costly and time consuming

(Huang et al 2003) Therefore, it is critical to keep

them under control Engineering changes have also

been recognised as a source of innovation and

creativity that can facilitate evolutions of products

and technologies (Balogun and Jenkins 2003, Jarratt

et al 2003, Eckert et al 2004) From this perspective,

knowledge acquired from engineering changes is

also very useful to product development in the long

term Despite the different perspectives, both of

the two arguments reflect the importance of

engineer-ing change management (ECM) in product

development

In the past, a lot of research have been done in

ECM regarding computerising traditional paper-based

engineering change processes (Huang et al 2001),

improving communicating methods between engineers

(Shiau and Wee 2008), clarifying knock-on changeeffects between components (Clarkson et al 2004,Eckert et al 2006) Most early research shows theefforts made in dealing with engineering changes in themanufacturing process Recently, changes happening

in the critical stage of product development, theproduct design stage, have been emphasised (Mckay

et al 2003) It is recognised that the design stage ofproduct development could determine the largest costsavings during the product life cycle This means,changes happening at the design stage would have agreater impact than those happening in the manufac-turing phase This project focuses on analysis ofchanges and their propagations in the product designphase and solving design conflicts arising from them byusing knowledge management technologies

As an important part of product development,ECM has been studied by many academia andindustrial practitioners in the past decade They haveidentified issues within ECM and tackled them withproposed solutions from different perspectives Anintroduction of the concept of ECM and an overview

of previous research are given below

*Corresponding author Email: j.gao@gre.ac.uk

International Journal of Computer Integrated Manufacturing

Vol 24, No 6, June 2011, 535–551

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

Ó 2011 Taylor & Francis

DOI: 10.1080/0951192X.2011.562544

1.1 The concept of engineering change management

Although engineering change has been studied for

many years, its definition varies according to

state-ments of different researchers Wright (1997) defined

engineering change as modification to a component of

a product before it goes into production Some

researchers agree that engineering change is

modifica-tion to dimensions, fits, forms, funcmodifica-tions and materials

to product or components after the product design is

released (Huang et al 2003, Kocar and Akgunduz

2010) Whilst some other researchers view engineering

changes as changes that occur in a wider range from

customer requirements to product in use (Pikosz and

Malmqvist 1998, Eckert et al 2004)

While research focuses have been shifting overtime,

the scope of research on engineering change has been

widened The initial motive of studying ECM was to avoid

engineering changes during the manufacturing process

due to the adverse effects they cause The adverse effects

caused in terms of delivery time, developing cost and

product quality are noticeable but very difficult to estimate

(Huang et al 2003) Later on, people realised that

engineering changes are actually inevitable Therefore,

researchers have turned to finding out how engineering

changes go on and what kind of impacts they may cause

(Clarkson et al 2004, Ouertani 2008, Kocar and

Akgunduz 2010) Recently, some researchers argue the

benefit of engineering changes to innovation and

creativ-ity, which can enhance the competitiveness of companies

Thus, some researchers have started to study engineering

changes from perspectives of knowledge management and

knowledge reuse (Balogun and Jenkins 2003, Palani

Rajan et al 2005, Lee et al 2006, Keese et al 2009)

The process of organising engineering change

activities has also been explored in the past decade,

in order to find most efficient and effective approach of

ECM A general process of engineering change has

been proposed by Clarkson and Eckert (2005) This

process includes six steps, namely engineering change

request, possible solution identification, risk/impact

assessment, solution selection and approval, solution

implementation and change process review Although

the general process of engineering describes a

reason-able approach to addressing change issues in product

development, in reality, different companies have quite

different processes to deal with engineering changes in

order to fit their specific organisational and production

requirements (Pikosz and Malmqvist 1998, Huang

et al 2003, Eckert et al 2004)

1.2 Change propagation analysis

One of the most difficult issues in analysing engineering

change is that when a component is changed, it may

also change its related components (Ariyo et al 2009).Therefore, an initial change may cause changesspreading at several structural levels Essentially, thereason why changes propagate is because of thedependency between components This situation isso-called change propagation or the knock-on effects

of changes, which makes change analysis very tricky.Some researchers have made some efforts in dealingwith this issue

Clarkson et al have proposed a method calledchange prediction matrix (CMP) to trace changepropagations and analyse the impacts they may cause(Clarkson et al 2004) The method transforms thedependency relationship between components in aproduct model to a design matrix Based on thismatrix, the likelihoods that potential change propaga-tions may happen between components are estimated.Also in the same way, the impacts these potentialchange propagations may cause have also beenestimated By combining the change, likelihood matrixand the change impact matrix, a change risk matrix hasbeen generated With the help of visualising method,change propagation paths and their relative risks havebeen clarified

In another study, Eckert et al (2004) haveproposed a method to analyse change propagation at

a parametric level and identify four types of changepropagation behaviours, namely constants, absorbers,carriers and multipliers These four types of changepropagation behaviours help to analyse change pro-pagations that cross multi-levels Four types of changepropagation behaviours represent four situations when

a change of a component propagating to anothercomponent via some other components, which in-cludes changes being passed without effect, beingreduced or eliminated, being replaced with newchanges from the intermediate component and beingenhanced This method has also been integrated withthe CMP method to enhance the performance ofchange propagation analysis in product conceptualdesign (Keller et al 2009)

Kocar and Akgunduz (2010) have proposed adifferent method to analyse change propagations Theyuse visualisation technique and data mining technique

to represent product models and find out dependenciesbetween components Users would be warned visually

if potential change propagation is predicted to happen.Ouertani (2008) has also proposed a visualisationtool called DEPNET to model product data and theirdependencies within them By using the product datadependent relationships, changes emerging duringcollaborative design process would be tracked down

Do et al (2008) have also proposed a method fortracking engineering change propagation betweendifferent product data views based on a shared base

product model The method reduces data redundancy

in ECM and maintains consistency between different

product data views

1.3 Information systems for ECM

Although methods for ECM have been proposed with

rigorously analytical or reasoning approaches,

de-signers will be easily exhausted before having

investi-gated the entire search space (Ariyo et al 2009)

Therefore, it is important that computer-aided tools

have been developed and used in ECM Previous

investigations have shown that computer-aided ECM

systems have been rarely utilised in companies (Pikosz

and Malmqvist 1998, Huang et al 2003) Although

some companies have used electronic document

man-agement systems to replace paper-based ECM

docu-ments, the data are non-structural and it is difficult to

semantically trace similar engineering change cases

Although currently not many companies are using

computer-aided systems to facilitate their ECM, there

are some systems that have been developed by academia

trying to enhance communication and information

sharing in change management process Huang et al

(2001) developed a web-based system to implement the

whole process of ECM, including engineering change

log, engineering change request, engineering change

evaluation and engineering change notice The

distrib-uted system has improved the efficiency of ECM and

enhanced the collaborations between engineers Also,

structural ECM data make it possible to integrate with

other computer-aided systems such as product data

management (PDM), enterprise resource planning

(ERP), computer-aided design (CAD) and supply chain

management (SCM) Ouertani and Gzara (2008) have

developed a system called DEPNET to visually track

dependencies within product specifications, so that

change propagations can be captured if any design

changes of a product specification happen As

men-tioned above, Kocar and Akgunduz (2010) developed a

visualisation system to track change propagation,

which is well integrated with 3D modelling system

Lee et al (2006) developed a knowledge-based system

to facilitate ECM in a collaborative environment The

authors have used ontology technology and case-based

reasoning method to construct a knowledge base of

previous development experience It also implements

the knowledge base with a web-based system that

enables users go through the whole ECM process from

change request initiated to change approved

1.4 Aim and objectives of the project

Based on previous research reviewed by the authors,

some gaps in ECM have been identified First, changes

of functional requirements should be consideredtogether with changes in the physical domain in thedesign phase For example, in the domain definitions

of the product design stage in the theory of AxiomaticDesign (Suh 2001), when the changes in physicalcomponents are considered, the changes in the func-tional domain and their effects on the physical domainhave not been considered Second, there is a lack ofconsideration of the impact of change solutions on thechange propagation analysis A lot of efforts have beenmade to predict change propagations with predefinedcomponent interactions However, specific solutionsfor change requests may dramatically change prede-fined interacting relationships between components,which may make predictions of later changes fail.Third, there is a lack of tools to help engineeringdesigners reuse knowledge from previous cases regard-ing design change management in industry In manydesign change cases, technical solutions for a designchange request, or say for some similar design changerequests, may have been re-developed on many otheroccasions That may be because experience or technicalsolutions from previous design change cases have notbeen formalised and shared effectively

This project therefore aims at dealing with changes

in the functional domain and the physical structuredomain of complex product design It would helpdesigners analyse design change propagations withinthese two domains It would also help to solve designconflicts arising from design changes by reusingknowledge from previous design change cases Thereare mainly three objectives, i.e (1) dynamically capturechanges and their propagations between functionalrequirements and physical structures; (2) identify andformalise design conflicts arising from design changesand (3) use knowledge-based engineering technology tofacilitate finding solutions for design conflicts fromprevious design cases A method for design changemanagement is proposed by putting the emphasis onanalysis of changes in the functional requirementdomain and the physical structure domain A model-ling method is employed to enhance the traceability ofchanges occurring between functional model andstructural model A matrix-based method is con-structed to capture dynamic change propagationsbetween the two domains In the end, a knowledge-based method is developed to help to solve conflictsarising during design change analysis An industrialexample has been used to show how the method works

2 Analysis of design change management (DCM)practices

Referring to the literature reviewed above, mostresearchers focus on the engineering change analysesInternational Journal of Computer Integrated Manufacturing 537

among physical components Although some of them

mentioned functional requirements would be

influ-enced by changes of physical components, detailed

discussion has not been made regarding how these

influences happen and how to deal with them In this

section, changes taking place between the functional

domain and the physical structure domain of the

product design stage have been discussed

Addition-ally, knowledge use during design change management

at this stage has also been covered

2.1 Change of functional requirements

Change of functional requirements may have many

reasons, for example, changes of customer demands,

changes of government policies or changes of project

aims for better competing with rivals (Rouibah and

Caskey 2003, Clarkson et al 2004) Changes

occur-ring in functional requirements may affect three

aspects of product design (1) Functional requirement

change needs to be verified according to customer

requirements to make sure all the changes meet the

original customer demands As one of the most

important inputs of a product design project,

customer requirements should be monitored all the

time during changes of functional requirements to

make sure all the changes that meet the original

customer demands (2) Any change of a functional

requirement may result in potential changes of other

functional requirements depending on the

interrela-tionships among them These changes will be

captured in the functional requirement model, so

that causal impacts can be analysed and controlled

(3) Obviously, any changes in the functional

require-ment domain will affect physical structures that are

correspondingly constructed according to functional

requirements

2.2 Change of physical structuresChange of physical structures is another importantpart of design change management A lot of situationsmay give rise to changes in the physical structuredesign, for example, changes of functional requirement(discussed above), physical conflicts within solutions,solution changes on the supplier’s side, technicalinnovation and manufacturing restrictions Changes

of physical structure may also directly affect threedomains, namely the functional requirement domain,the physical structure domain itself and the manufac-turing domain In the functional requirement domain,any changes in the physical structure may changetarget outputs of related functions Since one compo-nent is possibly involved in realisations of more thanone function, the relationships between componentsand related functions need to be clarified Therefore,any change in the physical domain needs be verified tomake sure that target functional requirements havebeen met In the physical structure domain itself,components are linked together by physical connec-tions, which make it possible to realise demandedfunctions Change of a component may potentiallychange operations of other components, which in turnmay change the realisation of related functions.Figure 1 shows the change propagation routes withinthe functional domain and the physical structuredomain and the routes between them For example,change of the functional requirement Fn may requirechanges on components involved in the realisation of

it, in this example let us say C2 The change of C2couldhave influence on Ci via some behavioural or spatialrelationships between them Ci is one of the compo-nents involved in the realisation of Fi Then Cineeds to

be checked against functional requirement Fito see if

Fican be satisfied If not, then Cior other components

Figure 1 Design changes between functional and physical domains

involved the realisation of Fimay need to be changed

to accommodate the influence of the previous change

These components may be also involved in realisations

of other functional requirements (for example F1, F2in

the diagram), so further changes may be needed In

this diagram, change of Fndirectly cause changes of its

components and also indirectly cause changes of other

components and functional requirements In the

manufacturing domain, change of physical structure

may change downstream activities, such as

manufac-turing process planning, SCM, risk and cost

evalua-tion Impacts on these product development activities

need to be analysed or re-evaluated Although change

propagation between the functional domain and

physical domain looks obvious from discussions in

the last two sections, there is a lack of method for

supporting tackling the change propagation process

between the two domains, solving problems arising

from the process and analysing change impacts An

important part of this article is focused on developing

such a method to bridge the gap The matrix-based

method for analysing change propagations has been

described in section 3

2.3 Conflict solving in design change management

Conflict solving is one of the most concerned issues in

design change management In many cases, changes of

a component or a function may require other parts of

the design to change correspondingly Furthermore,

changes of these parts would cause changes of more

parts of the design This effect is the so called change

propagation or knock-on effect Actually, the reason

why a change of a part of the design causes changes of

other parts is because the initial change of the design

may harm or obstruct operations of other components

or satisfactions of other functional targets, which can

be seen as functional or structural conflicts In other

words, change propagations are caused by design

conflicts Once there is no design conflict arising from

any design changes, the change propagation stops

Although many design conflicts may have been solved

during the change implementation by experienced

engineers, many others may not be recognised in the

design phase due to the lack of systematic methods and

they would have been carried over to the

manufactur-ing phase, which may cause a huge amount of cost in

later phases Therefore, an effectively analytical

method is needed to identify conflicts in design change

management and solve them as early as possible

2.4 Knowledge use in design change management

Knowledge use is critical for design change

manage-ment to ensure results from change analysis are fact

based and consistent There are mainly three aspectswhere design knowledge can be used to help solvechange problems

The first aspect is to identify design change modes.The design change mode is structured records of designchanges implemented in previous applications Thepoint of having design change modes formalised in aknowledge repository is for frontline design engineers

to find out whether similar design changes havehappened They can use these similar design changecases (if there is any) as references to help to findproper solutions and estimate potential change propa-gations and their impacts

The second aspect is regarding design conflictsolving During a company’s daily operations, thereare a lot of solutions that have been proposed byengineers in attempt to tackle design conflicts arisingfrom product development These solutions whethersuccessfully implemented or just on sketches areimportant assets of the company which should beproperly generalised and deposited in the knowledgerepository of the company Once new design conflictsemerge and there is no similar design change mode thatcan be referred to, engineers can follow a formalisedroute to try to find proper solutions for them

The third aspect is to use design knowledge tofacilitate change impact analysis Some knowledge ofphysical structure development has always beenstudied by companies, for example knowledge regard-ing developing time, developing cost and developingrisks of solutions, components and parts When adesign change is initiated, engineers not only need tofind its solutions and solutions for propagatingchanges, but also have to estimate the overall impactcaused by the initial design change by taking con-sideration of time, costs and risks for development ofnew solutions Therefore, decisions can be made forwhether it is worth proceeding or not, or which parts

of these solutions need to be modified, in order tomake sure the change impact will not be too heavy toafford

3 The proposed methodology3.1 Analysis of the DCM processThere are three types of relationships existing in aproduct design, i.e mapping relationship betweenfunctional requirements and physical structures, phy-sical interaction relationship between structures, andspatial connection relationship between structures(Christophe et al 2010) These relationships withinproduct design largely cause change propagations Themethod of design change management proposed in thisarticle is based on analyses of these three types ofrelationships

International Journal of Computer Integrated Manufacturing 539

The process of design change management has

been depicted in Figure 2 The whole process can be

divided into three main steps: (i) system modelling of

the product design; (ii) change propagation analysis

based on the composite matrix and (iii)

knowledge-based design conflict solving

In the system modelling step, the product design

with change (initial or propagated) applied has been

modelled by three types of models, i.e the functional

structure model (in the form of SysML block definition

diagram), the physical interaction model (in the form

of SysML activity diagram) and the spatial connection

model (in the form of CAD model) Each model

clarifies one aspect of the product design ingly By synthesising the relationships obtained fromthese three models, a composite matrix has beengenerated, which is critical to change propagationanalysis Differing from other matrix-based methods insome research in change propagation analysis, thecomposite matrix combines three types of relationships

correspond-of the design together and provides an intuitive anddynamic way to capture change impacts acrosscomponents and their functions Explanation of eachstep is given in Section 3.3

In the change propagation analysis step, thedesign change has been examined by checking

Figure 2 Analytical process of design change management

changes to related flows (representing interactions

between components) and spatial connections Any

components connected with these flows and spatial

connections are then examined with functions that

they serve for If the effects on the components, which

are caused by changes of those flows and spatial

connections, make realisations of corresponding

functions fail, it means there are design conflicts

existing, which are caused by the design change

Design conflicts are then identified during the analysis

with the composite matrix The process of change

propagation analysis using the composite matrix is

described in detail in section 3.4

In the knowledge-based design conflict solving step,

design knowledge which is acquired from previous

design cases is used to help solve design conflicts

identified in the last step Firstly, design conflicts are

formalised by using pre-defined engineering ontology

Secondly, the formalised design conflicts are reasoned

in the knowledge repository by semantically

compar-ing with formalised general design cases stored in the

knowledge repository General solutions in the

knowl-edge repository are formalised by the same set of the

pre-defined engineering ontology Thirdly, a prioritised

list is generated with the most semantically similar

general solutions at the top and the designers check

those general solutions starting from the top of the list

and their related design cases to find reference

solutions for the design conflicts At last, based on

the reference solutions, proper solutions are worked

out by designers and change decisions are made

The design with changes generated in this step will

be re-modelled and further possible change

propaga-tion are analysed again as what is done in the last two

steps

3.2 The industrial example used

The industrial example used in this project to evaluate

the proposed methodology is a cooling system, which

is a critical part of a wind turbine (Figure 3) This new

model of wind turbine is under development in our

collaborative company that is a pioneer in gearless

wind turbine development Nearly 2000 wind turbines

(by May 2010) based on their solutions have been

deployed in Europe, Asia and America There is a real

design change scenario that the wind turbine needs to

be deployed in a very sandy environment so that air

filtering measure of the cooling system needs to be

suppressed to prevent more sands than normal from

coming to the cooling system and damaging it This

change causes some knock-on effects that give rise to

changes on other parts of the system The

methodol-ogy proposed in this article is going to be used to solve

this problem by modelling the cooling system,

identifying design changes and related design conflictsand solving design conflicts by using a knowledge-based system

3.3 Modelling methods used in this projectModelling of engineering design includes three parts,namely functional structure model, physical interac-tion model and physical structure model In thisarticle, modelling methods for functional structureand physical interaction are adopted from SysMLTM,which is a comprehensive system engineering model-ling language (Object Management Group 2008) Thereason of using SysML is because it is a standardmodelling method having intuitively visual presenta-tions, standard descriptive language and software toolsupport It can be easily understood by both humanand computer, which is important for this project,since the methodology needs to be computerised toenhance its usability For this reason, SysML is betterthan other modelling methods Modelling of physicalstructure can be carried out by CAD systems, whichwill not be an emphasis in this article

The functional structure is modelled by the blockdefinition diagram (BDD) of SysML (Figure 4 depictsthe functional structure of the cooling system) TheBDD is used to model the hierarchically structuralrelationship of functions It also helps to clarify thespecifications of each function The specificationattribute of a function quantitatively or qualitativelyrepresents what the function has to do, which isanalysed by engineer from initial customer require-ments or other requirements from various sources (e.g.technical restriction, management and government).Specifications are represented as attribute-value (could

be precise value, value range or qualitative tion) pairs All of the sub-functions need to perform toFigure 3 An inside view of a wind turbine

descrip-International Journal of Computer Integrated Manufacturing 541

meet their corresponding specifications so that

speci-fications of their parent function can be met Any

changes to components need to be verified against its

corresponding functional specifications to check

whether these changes affect the realisation of

func-tions If functional specifications cannot be satisfied

due to these changes, then other consequent changes

need to be carried out

Physical structure modelling carried out by CAD

systems is intended to clarify the spatial relationship

between components When changes to a component

happen, the spatial relationship helps designers find

potential changes to neighbouring components based

on their positions The spatial relationship concerned

in this model is all about static or kinematic tions between components, which is based on assemblyrelationships, but does not involve any flow-basedphysical interactions

connec-Physical interaction relationship is modelled by theinternal block diagram to clarify the behaviouralrelationship between components Figure 5 shows theinteraction model of the cooling system There are avariety of flows going through components, includingenergy flows, material flows and signal flows A change

on a component may cause changes of the flows goingthrough it, which may also cause changes of upstream

Figure 4 Functional analysis of cooling system

Figure 5 Analysis of interactions in the cooling system

and downstream components involved in these flows.

That is because the status changes of these flows may

result in components not satisfying their corresponding

functional requirements

In order to be computerised, matrix analysis is

employed to represent change propagations within

and between functional requirement domain and

physical structure domain A composite matrix is

constructed based on the results of modelling

analyses of functional structure, physical interaction

and physically spatial relationship (Figure 6) The

matrix is composed of three parts that are marked

by different colours The first part (the green part)

represents the mapping relationship between function

and components Elements in the first column

represent physical components and elements in the

green part of the first row represent functions Each

marked cell in the green part represents the

involvement of a component in the realisation of a

function The second part (the blue part) represents

the interaction relationship between components,

which reflects the modelling results of physical

interactions between components Elements in the

blue part of the first row represent flows in Figure 5

When there are changes happening on a component

(an element in the first column), related flows going

through it will be identified in the matrix

Compo-nents that these flows go through are also identified

The third part (the grey part) represents the

physically spatial relationship between components

A marked cell in this part, the matrix means the

component in the column is physically connected

with the component in the row Clarification of this

type of relationship helps designers find potential

propagating changes to neighbouring components.The next section presents further explanation of howpotential propagating changes are identified

3.4 Identifying change propagation and designconflicts

Change analysis is intended to uncover changes andtheir propagations by following connections withinfunctional requirements and physical components andrelationships between them This section shows the idea

of identifying change propagations and their impactsarising from a change of a component The description

of the method is associated with a scenario ofimproving air filtering as mentioned above and based

on the composite matrix of change analysis (Figure 6).The process of identifying change propagation isdescribed in the following steps In this scenario, thechange is triggered by a functional requirement called

‘F2: Filter hot air’ Therefore, the analytical processstarts from the function-component part of the matrix(the green part)

Step 1: Identify component changes caused byfunctional change As mentioned above, because of thesandy environment where the wind turbine will bedeployed, the current air filtering measure cannot meetthe new functional requirement In Figure 6, it showscomponents involved in the realisation of function,filter hot air (F2) In this case, there is just onecomponent (C2, air filter mat) identified To meet thesandy environment, the current air filter mat with adust holding capacity 650 g/m2 needs to be changed

to a more effective one with dust holding capacity

750 g/m2

Figure 6 Composite matrix for change analysis

International Journal of Computer Integrated Manufacturing 543

Step 2: Identify potential affected components.

The component changed in the above step may

change the physical statuses of flows going through

it and also it may change its neighbouring

compo-nents due to changes of its spatial characteristics

Led by component C2, the row shows flows and

neighbouring components that are potentially

af-fected by the change of C2 In this case, flow FL1

(air from the generator) and neighbouring

compo-nent C1 (inner air incoming pipe) are related to C1

The flow FL1 also goes through C1, C3, C4, C5, so

these four components may also be potentially

affected by the change of C1 The side effects of

changing the air filter mat is that the mat with

higher dust holding capacity is thicker and it causes

larger air pressure drop, which can significantly

reduce the efficiency of heat exchanging

Step 3: Check change effects with related

func-tions Components that are affected by the flows and

the spatial connections need to be checked whether the

changed flows or the changed spatial connections

would affect the realisations of their related functions

In this case, the air flow after the filter mat has a lower

pressure, which means components C3, C4 and C5

would be potentially affected since the status of air

through them is changed (see the column led by FL1)

According to the analysis by engineers, the lower air

pressure through C3 (inner fan) will weaken its

performance Also the lower air pressure through C4

(air heat exchanger) will reduce the efficiency of heat

exchange But it has almost no effect on C5 (the inner

air outgoing pipe) The spatial change (thicker filter

mat) has been considered as not noticeable change on

C1 (inner air incoming pipe), since the change can be

easily accommodated by the current design Although

in this case change caused by spatial connection is

negligible, in many other cases, it may be significant

and corresponding changes need to be made For

example, in this case, change the component C1 or add

some other structures to accommodate the changes

caused by spatial connections Therefore, in this case,

C3 and C4 have been identified as affected components

which need to be changed to accommodate the

previous change on C2

Step 4: Identify and solve design conflicts By

analysing affected components, design conflicts can be

identified Taking C4 as an example, the changed input

flow is the incoming air pressure that is lowered and the

affected parameter is the heat exchange efficiency which

is also lowered The effect means that the heat exchange

cannot meet the functional requirement F4 Therefore,

this design conflict needs to be solved In this article, we

develop a knowledge-based method to help designers

find reference solutions from previous design cases

Detailed discussion of how to solve design conflicts

using a knowledge-based method is presented in section3.3

Step 5: Analyse change propagations caused bycomponent changes in step 4 When a proper solutionhas been found in step 4, changes on affectedcomponents have been determined These changeswould potentially affect other components as well Inthe above case, if component C4 has been changedflows FL1, FL2 and connected components C2, C6may also be potentially affected Thus, a next round ofchange analysis also needs to be carried out until there

is no further change effect being identified, whichmeans change propagation stops and change analysisinitiated by the first change is finished

4 Knowledge system support for design conflict solving

As the authors argued above, the reason why designchanges propagate is that there are design conflictsbetween components when one or some of themchanged TRIZ, which is originated from Russia, is aset effective problem solving methods (Altshuller 1996).The contradiction matrix and invention principles areuseful tools of TRIZ for solving technical conflicts Theidea of TRIZ to solve conflicts includes generally foursteps: (1) identify technical conflicts; (2) generalisetechnical conflicts by using 39 engineering parameters;(3) find invention principles via a standard contra-diction matrix and (4) explore specific solutions byfollowing the indications of invention principles (Alt-shuller 1996, Fey and Rivin 2005) Although techniques

of problem solving of TRIZ are innovative andinspirational to engineers, the method is difficult tomaster without a lot of trainings and long-timeexperience

In this article, the authors proposed a based method working in a similar way but moreintuitive and easier to use When a specific designconflict is identified during the change analysis, it will

knowledge-be formalised by referring to a set of predefinedontology Then the formalised design conflict will bereasoned in the knowledge repository to find semanti-cally similar generalised design cases Therefore,specific design cases that are associated with general-ised design cases can be retrieved These selected designcases will be used as reference solutions for the currentdesign conflict Figure 7 depicts the process of howdesign conflicts have been solved

4.1 Formalisation of design conflicts4.1.1 Design conflict

Design conflicts are identified in the change tion analysis based on the composite matrix

Occurrence of design conflict has been simply depicted

in Figure 8 in order to help understand the idea Given

that component 2 is one of the components serving a

function, when there is a change request applied to a

component (component 1), which has interactional

connection with component 2, it may change the input

flow of component 2, which may further affect its

output flow If the affected output flow cannot satisfy

the requirement of the function, then it is said that

there is design conflict occurring at component 2,

which is caused by the previous change request to

component 1

4.1.2 Formalising design conflict with predefined

ontology

Formalising a design conflict is actually not

formalis-ing the design conflict itself In fact, it is about

formalising the interactional model (called the

meta-interactional model) of the component where the

design conflict occurs Figure 9 shows formalisation

of a meta-interaction model

A meta-interaction model includes a physical

component where the conflict happens, the changed

input flows and affected output flows Both the input

flows and the output flows are formalised by the flow

ontology and characteristics ontology (ontology picted in Figure 10) The flow ontology defines the type

de-of the flow The characteristics ontology defines theproperties of the flow For example, the gas flownormally has properties like pressure, temperature,moisture, etc Properties formalised in this part should

be critical to the operation of the component Concepts

of the characteristics ontology (a node of the ontologystructure) are associated with concepts of the flowontology in the form of their properties (shown inFigure 10)

The component is also formalised by the viour ontology and the component ontology Thebehaviour ontology defines what the component doeswith the input flows and generates the output flows.The component ontology defines which type ofcomponent it is The component ontology containsrelated component characteristics as its properties.These component characteristics are critical for theperformance of the operation of the component Forexample, in the cooling system, there are twocharacteristics of the heat exchanger that areimportant to its functionality One is the area ofthe heat exchanger surface Wider surface can have ahigher heat exchanging efficiency The other one isthe material that the heat exchanger is made of

beha-Figure 7 Process of solving design conflicts

Figure 8 Design conflict occurring

International Journal of Computer Integrated Manufacturing 545

Some material, for example bronze, has a better heat

conduction performance than some others, for

example steel

The flow ontology and the behaviour ontologyused in this article are adopted from the work of Hirtz

et al (2002) They proposed a functional ontology, so

Figure 9 Formalisation of the meta-interaction model

Figure 10 Ontology development for design change management

called functional basis, contains generalised functions

(called as behaviour in this paper to differ from the

term function of design) and flows which are seen as a

useful and comprehensive engineering functional

ontology by the authors Also the authors developed

domain-based component ontology and its

character-istics by synthesising terms used in the investigated

company The characteristics ontology associated with

flow ontology is also developed by synthesising related

work by other researchers We adopt Prote´ge´ as an

ontology editor to formalise predefined ontologies in

computer language The tool which is developed by

Stanford University is de facto in the academic area for

ontology development

Figure 10 shows some parts of the ontology,

including definitions of flows, definitions of behaviours

and definitions of components

4.2 The knowledge system for design conflict solving

Figure 11 shows the framework of the knowledge base

for design conflict solving When a design conflict

arises from design change analysis, it is formalised in

the way as discussed in section 4.1 The formalised

model of the design conflict will then be thrown into

the knowledge system The system will reason in the

knowledge repository by analysing the semantic

similarities between different concepts to find most

similar generalised design cases After that, design

cases associated to these generalised design cases will

be retrieved as reference solutions for the current

design conflicts Designers can adjust or adopt

retrieved reference solutions to solve current problems

The method of generalising design cases is as the same

as the way formalising meta-interaction model It

collects design cases and formalises their target

problems or functions The generalised design caseswork as indices of those associated physical designcases

In this system, design cases are collected from manysources including different functional departments and

IT systems The design cases are formalised in ahierarchical way, which clarifies functions or problems

a design case is to address, the solutions used in thisdesign case, the components involved in this solutionand characteristics that contribute to the realisation ofthe function or the problem solving The formalisedstructure is also been generalised by the predefinedontology

4.3 Reasoning method for design conflict resolvingThe reasoning method is critical to finding referencesolutions for design conflicts arising from designchange propagation analysis The general approach

to solving design conflicts is depicted in Figure 12

In the reasoning method, one of the mostimportant steps is to analyse semantic similaritiesbetween formalised design conflicts and generaliseddesign cases stored in the knowledge repository.Given that a design conflict is defined as a definition(D) and all of the stored general design cases aredefined as a set of definitions {D1 Dn}, the firststep is to find the most similar solutions from the set

of general design cases The algorithm of calculatingthe semantic similarities between a formalised designconflict and a generalised design case is comparingthe semantic similarity of each corresponding elementand then adding them up to get an overall semanticsimilarity

Figure 13a represents hierarchically ontologicaldefinition of a group of concepts The higher of levels

a concept stays in, the more general semantic

Figure 11 Framework of the knowledge system for conflict solving

International Journal of Computer Integrated Manufacturing 547

meaning it represents While in lower levels, the

semantic meaning of a concept is more specific In

Figure 13b, IC (S1) represents the semantic meaning

of the concept S1 Since S1 is the parent of S11 and

S12, S1 has a wider semantic meaning than S11 and

S12, which means:

ICðS11Þ 2 ICðS1Þ; and ICðS12Þ 2 ICðS1Þ ð1Þ

Theoretically, the following equation can represent

how S11 (a child) is semantically similar to S1 (a

parent) by comparing scales of semantic meaning of

each concept:

SimðS11;S1Þ ¼ ICðS11Þ=ICðS1Þ ð2Þ

While the similarity of S11 (a brother) to S12

(a brother) can be expressed as:

SimðS11; S12Þ ¼ ðICðS11Þ \ ICðS12ÞÞ=ICðS12Þ ð3Þ

While in reality, it is well-known that exact

semantic meaning of a concept is very difficult to

measure More often than not, people can tell the

qualitative similarity between two concepts by theirexperience and common sense Particularly, people inthe professional area are better to tell similaritiesbetween concepts of professional terms some of whichare not normally used by people out of the area So inthis research, a survey is developed and experiencedengineers are interviewed to rate semantic similaritiesbetween child-concepts and their direct parent-con-cepts (e.g Sim(S11, S1)) and also between theirbrother-concepts (e.g Sim(S11, S12)) The survey andthe rating process is a multi-criteria decision-making process, which is not discussed in details inthis article

With rated similarities between parent–child cepts and brother concepts, the similarity of any twoconcepts (e.g S111 and S22) can be expressed as:

con-SimðS111; S22Þ ¼

SimðS111; S11Þ  SimðS11; S1Þ

Figure 12 Reasoning approach to design conflict solving

Figure 13 Comparison of semantic meanings between concepts

formalised design conflict and a generalised design case

can be described as:

SimðDC; GDÞ ¼n

SimðCFDC;CFGDÞ

 Sim ðBehaviourDC;BehaviourGDÞ

 Sim ðComponentDC;ComponentGDÞ

 Sim ðAFDC;AFGDÞo

ð5Þ

Where DC represents design conflict, GD

repre-sents generalised design case, CF reprerepre-sents changed

flow and AF represents affected flow Similarity

between changed flows can be represented as:

SimðCFDC;CFGDÞ¼ Sim ðFlowDC;FlowGDÞ

 Sim ðCharacterDC;CharacterGDÞ

ð6Þ

Similarity between affected flows can be

repre-sented as:

SimðAFDC;AFGDÞ¼ Sim ðFlowDC;FlowGDÞ

Sim ðCharacterDC;CharacterGDÞ

ð7Þ

By comparing the overall similarities between

the formalised deign conflict and the generalised

design case, a set of prioritised similarity values are

generated:

fSimðDC; GD1Þ; SimðDC; GD2Þ; ; SimðDC; GDnÞg

By exploring and reviewing design cases associated

with retrieved generalised design cases from higher

priority to lower priority, the suitable design cases are

chosen as reference solutions for the target design

conflict

4.4 Results of the industrial example and system

implementation

The example used in previous sections has been

processed in this system Application of the

methodol-ogy has been partially displayed during the discussion

of the methodology In order to have an overview of

the results and the general process in an intuitive way,

a tabular form has been used to organise the results

(shown in Figure 14)

A three-tier web-based pilot system has also been

developed to implement the proposed methodology for

engineering design change management The pilotsystem is partly developed Integrations with otherbusiness systems like PDM, ERP and SCM are going

to be done at the next stage of this research Softwaresystems involved in this prototype include MySQL as

an infrastructure of data storage, Tomcat 6.0 as a webserver and servlet container, Java enterprise edition(Java EE) as the business implementation architectureand also JavaServer Page as a presentation technology.Figure 15 shows an overview interface of designconflict solving of design change analysis The pageshows the final stage of the design conflict solvinganalysis process, which is trying to get referencesolutions from the knowledge repository to helpdesigners work out a viable solution for the currentdesign conflicts

5 Conclusions and further workAny design changes either in functional requirementdomain or in the physical structure domain willpotentially affect operations of other parts Changepropagations and their impacts are difficult to becaptured, which makes product design in uncer-tainty The authors argue that design changepropagation is caused by design conflicts arisingfrom the initial change and changes afterwards.Change propagations are difficult to predict withoutknowing their preceding change solutions, since allfollowing changes are based on the solutions ofpreceding changes Knowledge from previous designchange cases is an important asset for companies.Many design conflicts arising from change analysiscan be tackled by reusing well-formalised and-managed knowledge abstracted from previous de-sign cases In this article, a methodology for designchange management has been proposed to implementthese ideas by using modelling method, matrixanalytical method and knowledge support First,the system engineering modelling method capturescritical interactions between functions and compo-nents The matrix-based analytical method helps totrace change propagations and identify design con-flicts by following mapping relationship betweenfunctions and components, and behavioural andspatial connections between components By usingthe design conflicts formalisation approach, designconflicts can be formalised based on predefinedontology, which makes knowledge reasoning in theknowledge base possible The framework of theknowledge base is proposed to collect knowledge ofdesign change from previous design cases With help

of the knowledge base and the reasoning method,designers are able to find proper solutions andevaluate their potential impacts A prototype systemInternational Journal of Computer Integrated Manufacturing 549

has been developed to show that the idea of this

article is technically feasible and can benefit

compa-nies with IT technologies

The emphasis of further work will be put on

development of methods for evaluating design change

impact in terms of factors of project success Any

changes made in product development need to be

evaluated in terms of development time, development

risk and development cost However, in practice,

impacts of design changes are normally estimated by

experienced engineers or other staff in the company

The results of change impact evaluations could be veryinconsistent with estimations from different peoplewho have different experience and different knowledgebackgrounds In the next stage of this research, anapproach to integrating with other enterprise systemswill be developed This approach is to acquire knowl-edge regarding contributions that previous productsmake to related projects in terms of factors of projectsuccess With help of knowledge acquired, an algo-rithm will be developed to calculate change impactmore accurately

Figure 15 Prototype system implementation of CDM

Figure 14 Results of the example processed in the knowledge system

This project has been supported by the wind turbine design

company Vensys Energy AG in Germany and the leading

wind turbine manufacturer Goldwind Ltd in China Some of

their engineering designers provided advice in developing the

methodology and the illustrative example

References

Altshuller, G.S., 1996 And suddenly the inventor appeared:

TRIZ, the theory of inventive problem solving 2nd ed

Worcester, MA: Technical Innovation Center

Ariyo, O.O., Eckert, C.M., and Clarkson, P.J., 2009

Challenges in identifying the knock-on effects of

en-gineering change International Journal of Design

En-gineering, 2, 414–431

Balogun, J and Jenkins, M., 2003 Re-conceiving change

management: a knowledge-based perspective European

Management Journal, 21 (2), 247–257

Christophe, F., Bernard, A., and Coatane´a, E´., 2010 RFBS:

a model for knowledge representation of conceptual

design CIRP Annals-Manufacturing Technology, 59 (1),

155–158

Clarkson, J and Eckert, C., 2005 Design process

improve-ment: a review of current practice London: Springer

Clarkson, P.J., Simons, C., and Eckert, C., 2004 Predicting

change propagation in complex design Journal of

Mechanical Design, 126 (5), 788–797

Do, N., Choi, I.J., and Song, M., 2008 Propagation of

engineering changes to multiple product data views using

history of product structure changes International

Journal of Computer Integrated Manufacturing, 21 (1),

19–32

Eckert, C., Clarkson, P.J., and Zanker, W., 2004 Change

and customisation in complex engineering domains

Research in Engineering Design, 15 (1), 1–21

Eckert, C.M., et al., 2006 Supporting change processes in

design: complexity, prediction and reliability Reliability

Engineering & System Safety, 91 (12), 1521–1534

Fey, V and Rivin, E.I., 2005 Innovation on demand: new

product development using TRIZ Cambridge: Cambridge

University Press

Hirtz, J., et al., 2002 A functional basis for engineering

design: reconciling and evolving previous efforts

Re-search in Engineering Design, 13 (2), 65–82

Huang, G.Q., Yee, W.Y., and Mak, K.L., 2001

Develop-ment of a web-based system for engineering change

management Robotics and Computer-Integrated

Manu-facturing, 17, 255–267

Huang, G.Q., Yee, W.Y., and Mak, K.L., 2003 Current

practice of engineering change management in Hong

Kong manufacturing industries Journal of Materials

Processing Technology, 139 (1–3), 481–487

Jarratt, T.A.W., et al., 2003 Environmental legislation as adriver of design In: A Folkeson, K Gralen, M Norell,and U Sellgren, eds The 14th international conference onengineering design (ICED’03), 19–21 August 2003 Stock-holm, Sweden

Keese, D.A., Seepersad, C.C., and Wood, K.L., 2009.Product flexibility measurement with enhanced changemodes and effects analysis CMEA International Journal

of Mass Customisation, 3, 115–145

Keller, R., Eckert, C.M., and Clarkson, P.J., 2009 Using anengineering change methodology to support conceptualdesign Journal of Engineering Design, 20 (6), 571–587.Kocar, V and Akgunduz, A., 2010 ADVICE: a virtualenvironment for engineering change management Com-puters in Industry, 61 (1), 15–28

Lee, H., et al., 2006 Capturing and reusing knowledge inengineering change management: a case of automobiledevelopment Information Systems Frontiers, 8 (5),375–394

Mckay, K., et al., 2003 Design change impact analysisduring early design specification International Journal ofComputer Integrated Manufacturing, 16 (7), 598–604.Object Management Group, 2008 OMG Systems ModelingLanguage (OMG SysML) V1.1 [online] Object Man-agement Group Available from: http://www.omg.org/spec/SysML/1.1/PDF [Accessed 16 May 2009]

Ouertani, M.Z., 2008 Supporting conflict management incollaborative design: an approach to assess engineeringchange impacts Computers in Industry, 59 (9), 882–893.Ouertani, M.Z and Gzara, L., 2008 Tracking productspecification dependencies in collaborative design forconflict management Computer-Aided Design, 40 (7),828–837

Palani Rajan, P.K., et al., 2005 An empirical foundation forproduct flexibility Design Studies, 26 (4), 405–438.Pikosz, P and Malmqvist, J., 1998 A comparative study ofengineering change management in three Swedish en-ginnering companies In: H Lipkin and F Mistree, eds.Design engineering technical conferences and computers inengineering conference, 13–16 September 1998 Atlanta,Georgia

Rouibah, K and Caskey, K.R., 2003 Change management

in concurrent engineering from a parameter perspective.Computers in Industry, 50 (1), 15–34

Shiau, J.-Y and Wee, H.M., 2008 A distributed changecontrol workflow for collaborative design network.Computers in Industry, 59 (2–3), 119–127

Suh, N.P., 2001 Axiomatic design: advances and applications.New York: Oxford University Press

Wright, I.C., 1997 A review of research into engineeringchange management: implications for product design.Design Studies, 18 (1), 33–42

International Journal of Computer Integrated Manufacturing 551

Weighted nested partitions based on differential evolution (WNPDE) algorithm-based scheduling of parallel batching processing machines (BPM) with incompatible families and dynamic lot arrival

Guojun Su and Xionghai Wang*

College of Electrical Engineering, Zhejiang University, Hangzhou, China(Received 9 September 2010; final version received 6 February 2011)The scheduling problem of parallel batching processing machines (BPM) with incompatible families and dynamic lotarrivals in the diffusion and oxidation of semiconductor production line is studied Given that the batch schedulingproblem is NP-hard, it is decomposed into three sub-problems: batch forming, machine assignment and batchsequencing A novel algorithm named weighted nested partitions based on differential evolution (WNPDE) isproposed for the most critical sub-problem: machine assignment sub-problem The WNPDE retains the globalperspective of nested partitions algorithm and the local search capabilities of differential evolution (DE) algorithm.Experimental results indicate that the proposed WNPDE algorithm has high efficiency to minimise the totalweighted tardiness (TWT)

Keywords: batch scheduling; nested partitions; differential evolution; total weighted tardiness

1 Introduction

It is a very challenging task in semiconductor wafer

fabrication industry to meet one of the most important

customer’s expectations: maximising on-time delivery

or minimising tardiness This research focuses on

scheduling of batching processing machines (BPM)

found in the diffusion and oxidation area of

semi-conductor production line (SPL) BPM can process a

batch simultaneously (a batch, which includes six to 12

lots, is a collection of lots that are processed at the

same time on the same machine) BPM operations may

take roughly of the order of 10 h as opposed to 1 or 2 h

for most other processes of semiconductor

manufac-ture facilities; therefore, the effective scheduling of

BPM is a key factor to improve the productivity and

system performance (Perez et al 2005) of SPL

Most of the literatures on batch scheduling about

BPM are divided into three categories: single

machine-compatible lot families (Gupta and Sivakumar 2006),

single machine-incompatible lot families (Kurz and

Mason 2008) and parallel machines-incompatible lot

families (Cheng et al 2008, Chiang et al 2010)

Apparent tardiness cost (ATC) rule and batch ATC

(BATC) rule were proposed by Vepsalainen and

Morton (1987) and Mehta and Uzsoy (1998); the

ATC rule was used to form the batches and the BATC

rule was used to assign the batches to machines The

ATC and BATC rules that took into account were

product weight, batch processing time, due date and

other factors that usually showed good performance

on the minimisation of total weighted tardiness(TWT) However, in real-world situations, it is useful

to consider the realistic situation that the lots haveunequal ready time To deal with the dynamic situation,the dynamic batch dispatching heuristic (DBDH) rulewas extended from the ATC-BATC rule by Ilka andLars (2003), and three modified heuristic rules based onthe DBDH were also proposed by Mo¨nch et al (2005)

Li et al (2009) illustrated the advantages to useinformation on future arrivals for making decisions in

a dynamic batch-processing environment Glassey andWeng (1991) introduced the concept of using futurearrival information and called it look-ahead Fowler

et al.(1992) expanded the concept to multiple productsand created the heuristic method called the next arrivalcontrol heuristic (NACH) NACH only considered thenext arrival and makes the decision whether to start thebatch now or wait for the next arrival The NACHþproposed by Ham and Fowler (2008) based on theNACH controlled the incoming inventory into batchoperation Later, Reichelt and Mo¨nch (2006) addressedthe problem with bi-objective including minimisingTWT and makespan at the same time based on theapproach (Mo¨nch et al 2005)

Although scheduling of BPM has been widelystudied, the researches considering the situationssimultaneously as parallel machines-incompatible lotfamilies and dynamic lot arrival are relatively less due

to complexity In this research, we expand the parallel

*Corresponding author Email: wxh_10@zju.edu.cn

International Journal of Computer Integrated Manufacturing

Vol 24, No 6, June 2011, 552–560

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

Ó 2011 Taylor & Francis

DOI: 10.1080/0951192X.2011.562545

machines-incompatible lot families batch scheduling

problems considered by Balasubramanian et al (2004)

to a more realistic case of lots with dynamic release

times, that is, the lots’ release times are not equal

Unequal release time makes the batch scheduling of

BPM more difficult

2 Problem assumptions and notations

Our reasonable assumptions are as follows:

(1) Lots of the same family have the same

(4) Once a machine is started, it cannot be

inter-rupted, i.e no pre-emption is allowed

(5) Lots fall into different incompatible families

that cannot be processed together

We use the following notation throughout the rest of

the article

(1) There are m identical machines in parallel

(2) There are nj lots of family, j waiting to be

scheduled at the moment t0

(3) The processing time of family j is represented as

pj

(4) The capacity of batch processing machine is B

lots

(5) Lot i of family j is represented as ij

(6) The weight of lot ij is represented as wij, the

arriving time of ij is represented as Rij, the

processing time of ij is represented as Pij,

the completion time of ij is represented

as Cij and the due date of lot ij is represented

as dij

(7) The weighted tardiness of lot ij is represented as

wij6 max(Tij,0), where Tij¼ Cij7dij is

tardi-ness of lot ij

The objective of batch scheduling is to minimise theTWT,

to the machines and sequencing the batches In mostcases, it is worthwhile to form a batch with higherdegree of fullness, so that the manufacturing capacity

is not wasted; however, a tendency to fill out the batchruns the risk of postponing the lots that arrive earlier

in order to wait for other lots that arrive relatively late.After the batches are formed, we need to assign formedbatches to a suitable machine; at last, the batchesassigned for each machines are sequenced properly.Batch forming and sequencing sub-problems areillustrated in Section 3.2, and the proposed weightednested partitions (NP) based on WNPDE is described

in detail in Section 4

As an example illustrated in Figure 1, there areeight dynamic arrival lots belonging to three families inthe time window interval (t, tþ Dt) Dt, is timewindow, and the batch size is assumed to be 2 Theeight lots are grouped into four batches by DBDH(sub-problem 1), then batches are assigned to threemachines by proposed WNPDE (sub-problem 2) and

at last, the batches of each machine are sequenced byDBDH (sub-problem 3)

3.2 Batch forming and sequencing

We form the batches using the time window concept(Mo¨nch et al 2005) as follows:

Dt¼ AvgðPijÞwhere Avg(Pij) is the average processing time of Lij

Figure 1 Example of scheduling of BPM

International Journal of Computer Integrated Manufacturing 553