An integrated process planning and robust fixture design system 1

Fixture design, one of the activities that involves both product design and manufacturing, refers to the method and device employed to locate, support and hold a part to be machined firm

Introduction

The use of computers has changed the way of engineering design and manufacturing

of products Among the recent trends in manufacturing are: products are having a shorter life-cycle, frequent design changes, and lower production cost with better product quality The integration and optimization of design with manufacturing is the logical way to realize such trends

Nowadays, many companies are relocating their operations offshore to reduce labour costs Workforce and machine resources have been de-centralized and re-distributed throughout the world With the expansion of computing networking and Internet which makes ubiquitous accessibility of data possible, there is a growing need to integrate and optimize design and manufacturing activities across the geographical boundaries among globally distributed resources, which will benefit the manufacturing industries significantly

Fixture design, one of the activities that involves both product design and manufacturing, refers to the method and device employed to locate, support and hold

a part to be machined firmly in a machine tool during a machining process Proper

fixture design is crucial to product quality in terms of meeting the final specification

of the machined part Fixturing systems, usually consisting of clamps, supports and locators, must be capable of positioning, holding, and supporting the workpiece throughout the machining process The experience of process planners indicates that the process plan for machining processes cannot be finalized until a workable scheme for fixture layout has been determined Since process planning is the key to the integration of design and manufacturing, fixture planning becomes a critical issue to

be resolved for the realization of integrated concurrent design and manufacturing activities

However, current product design methodology does not adequately recognize the constraints the product definition imposes on fixturability problems one would face during machining operations Traditionally, fixture designers in industry rely mainly

on past experience, extensive experimentation or trial-and-error methods to determine

an appropriate fixturing scheme Set-up planning, fixture element design, and fixture layout design are often dealt with at the downstream end of the product development life-cycle These practices do not lend themselves well to the bridging of design and manufacturing activities, and have been one of the main reasons for the long lead time

in product development which many industries are striving to reduce

This research addresses these issues Integrative features, i.e., machining features, fixturing features, set-up features and machine resource features, are used to integrate the set-up planning, machining resource planning and fixture planning from product conceptualization to optimizing the product and its manufacturing processes simultaneously to meet cost and performance objectives Thus product quality and

user satisfaction can be best achieved Ant colony optimization and genetic algorithm are employed for the optimization processes in this study

1.2 Background

1.2.1 Fixture Design in the Integration of Design and Manufacturing

Recently, many approaches have been used for the integration of design and manufacturing Design for manufacturing, simultaneous engineering, or concurrent product and process design are among these approaches which entail the concept of designing a product and its manufacturing processes simultaneously

An overview of current and past fixture design and planning research has been given

by Nee et al (2004) Classic efforts for integrating fixture planning with CAD/CAM

can be found in the work by Van’t Erve and Kals (1986a, b) where they reported the development of a knowledge-based generative process planning system named XPLANE (eXpert process PLANning Environment) XPLANE includes a link to CAD systems and the selection of jigs and fixtures, NC part-program generation, tool management and capacity planning Recently, Fathianathan (2004) developed a middleware to integrate design and manufacturing activities, using fixture design as

an application domain Other efforts reported especially in this area are Kumar et al (1992), Fuh et al (1993), Ong and Nee (1994, 1996, 1997, 1998), Shirinzadeh (1996),

and Smith (1996)

Some of the problems of existing integrated design, manufacturing and fixturing systems are:

1 It is explicitly or implicitly assumed in most research work that fixture planning is usually performed after process planning Most research work concentrates on the design and layout of a fixture for an individual machining step in a process plan The fact that fixture planning may conversely affect process planning and concurrent product design and fixture planning has not been studied in great details

2 There has been little or no clear understanding of the relationship and interaction between fixture planning, product design and process planning, i.e.,

no clear knowledge link between design and manufacturing and no efficient knowledge representation scheme This can be attributed to the difficulty of capturing design knowledge in fixture design and planning, as this subject is often considered to have as much craft as analytical content

3 Some factors having a significant effect on fixture planning have been neglected in much of the previous research These include, for example, the position and orientation tolerances and the form irregularity, or the unavailability of finished or unfinished locating surfaces, of the initial workpiece or raw part, such as typical cast and forged parts shown in Figure 1.1

4 Design and manufacturing integration requires automatic extraction and interpretation of design and manufacturing features from a CAD model with minimum human intervention However, most of research work has paid relatively little attention to fixturing features in fixture design, i.e., to comprehend and extract fixturing characteristics of a part

Figure 1.1 Typical cast and forged parts

5 Five-axis machine tools can offer manufacturers many advantages, including reduced set-up times, lower costs per part, more accurate machining, and improved part quality However, most of the previous work focused on fixture planning for 3-axis machining and there is very little reported work for 5-axis machining

6 Few of the systems reported consider the existing constraints in the real factory environment for the manufacture of a part

7 The issue addressing the robustness of fixture layout has not been studied intensively Although a number of studies have been reported to minimize localization errors, it does not guarantee a fixture layout solution to be least sensitive to the variation of localization errors Other than making the

localization errors as small as possible, to make the selected locating features less sensitive to the external errors such as locating surface errors, set-up errors and fixture errors which result in the localization errors is also vital

1.2.2 Feature-based Technology

Feature technology is the technology of using features in the representation schemes

of product development It is used to create or recognize features on mechanical components and use them for downstream applications It can be considered as the extension of a parametric CAD system Parametric CAD systems are used to design objects that represent a family of variational geometry While CAD data consists of only geometric and topologic information, features are used to represent an artefact that might be manufactured or assembled (application) In the most general sense, features are high-level entities that model the correspondence between design information and manufacturing activities (Regli, 1995) Feature technology is found suitable for the integration of CAx segments due to its ability to capture and carry the designer’s intent from one stage to the other of product development (Shah, 1990) Feature-based modelling systems support more levels of information of downstream applications to a higher degree Research efforts in the integration of various CAx segments using feature technology have been reported and literature is available

especially for the integration of design and manufacturing (Ito et al., 1988; Suzuki et

al 1988; Kingsly Jeba Singh, 2006) Shah (1991), Bronsvoort and Jansen (1993),

Salomons et al (1993) and Subrahmanyam and Wozny (1995), Han (1996) and Han et

al (2000) have reviewed the major concepts and approaches used by researchers in

feature technology

1.2.3 Genetic Algorithm

The Genetic Algorithm (GA), which was first developed by John Holland (1975), is a stochastic optimization technique based on the mechanics of natural selection and natural genetics This algorithm works on a population of designs The population evolves from generation to generation, gradually improving its adaptation to the environment through natural selection, and fitter individuals have better chances of transmitting their characteristics to the later generations GA combines the concept of survival of the fittest among solutions with a structured, yet randomized information exchange and offspring creation In the algorithm, the selection of the natural environment is replaced by artificial selection based on a computed fitness for each design The term fitness is used to designate the chromosome’s chances of survival, and it is essentially the objective function of the optimization problem The genetic algorithms have been proven to be a useful technique in solving optimization

problems in engineering (Goldberg, 1989)

1.2.4 Ant Colony Optimization

Ant Colony Optimization algorithm (ACO) is a probabilistic technique for solving computational problems, and it mimics ant activities in finding food in the real world

It is a meta-heuristic-like generic algorithm and has been used for solving many different discrete optimization problems Dorigo and Gamberdella (1997) proposed ACO and applied it to the traveling-salesman problem They also compared the solutions of ACO and showed it to be better than other heuristic approaches like GA, evolutionary programming (EP), simulated annealing (SA) and a combination of GA

and SA Dorigo et al (1992), Jayaraman et al (2000), and McMullan (2001) have

already proved that an ACO is a useful technique in solving optimization problems in

engineering applications It has been applied to optimize product design and

manufacturing processes For example, Basker et al (2004) evaluated the production

cost and production rate for the optimal grinding condition, subject to constraints such

as thermal damage, wheel wear parameters, machine tool stiffness, and surface finish, and they proved that the ant colony system has outperformed the quadratic

programming technique and the genetic algorithm

1.3 Scope of the Present Research

This research addresses the development of an integrated process planning and robust fixture design environment for cast and forged parts, and this integration simultaneously optimizes the product and its manufacturing processes to meet cost and performance objectives The task is achieved by considering the product design intent (function, aesthetics, etc) and fixture design, in terms of locating, clamping and maintaining various tolerance relationships, as an integrated process Integrative features, that is, machining features, fixturing features, setup features and machine resource features, are used as the link between the two functions of product design and manufacturing The definitions of them are made and the links between them are described Functional modules are used to embody product design, machine resource and process processing which may be distributed in geographically different places

A hybrid fixturing-feature-based approach is presented to include fixturing features of

a product at the concept product design stage Machining and fixturing features, which are the segments of the integrative features, are automatically extracted from

the 3D model of a concept product design initially If the extracted fixturing features are not sufficient, additional fixturing features can be designed using the design-by-fixturing-feature approach or be defined through the interactive fixturing feature definition approach The designer, however, need to bear in mind of the size, functionalities and aesthetics which may be altered by inclusion of fixturing features

In some cases, such features may have to be removed after the machining operations and the process will induce greater production cost and time The obtained machining/fixturing features are stored in an information structure which can be accessed by each segment of the product development cycle

It is assumed that a machining environment usually contains several machine resources which can be distributed and located in different places The application of fixturing features may vary according to different machining environment, e.g 3-, 4-

or 5-axes machining centre, vertical or horizontal, etc The ideal strategy used in applying fixturing features mentioned above is realized by planning the optimum scheduling of the machine resources This is achieved by minimizing a cost model among the distributed machine resources during the process planning

A robust fixture layout approach is also presented In this approach, a fixture layout is treated as a multi-objective optimization problem which considers both the mean localization error and the variation of the mean localization error of the machining features in a set-up as well as the location stability and reliability The sources of the localization errors are analyzed and the mathematical model to calculate the point-wise localization error is established The distance of the location points among the

locators is taken into account to ensure the location is stable and reliable Non-dominated GA and ACO are proposed to solve this multi-objective problem

1.4 Objectives of the Present Research

The objectives of the research are:

 To define the properties of integrative features and describe the knowledge

links between them Their properties and links shall ensure the integration of design and manufacturing

 To define the tasks of the four functional modules of product design, machine

resource, process planningand managing

 To use a hybrid feature-based approach to obtain fixturing features to integrate

the process planning and fixture design processes, which includes the approach of automatic feature extraction, design-by-fixturing-feature and interactive definition of fixturing features

 To optimize process planning to meet cost and performance objectives using

ACO and consider the selection of available machine tools, tolerance analysis and cost modelling simultaneously

 To optimize fixture layout by considering it as a three-objective optimization

problem and use non-dominated multi-objective ACO and GA for its solution