International journal of computer integrated manufacturing , tập 23, số 5, 2010

Integrated line balancing to attain Shojinka in a multiple straight line facilityHadi Go¨kc¸ena, Yakup Karab* and Yakup Atasagunba Department of Industrial Engineering, Faculty of Engine

Prevention of resource trading fraud in manufacturing grid: a signalling games approach

Haijun Zhanga,b*, Yefa Huaand Zude Zhouaa

School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China;bDepartment of

Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109-2125, USA

(Received 13 March 2009; final version received 14 January 2010)

In the manufacturing grid resource market, the information asymmetry for buyers and sellers is a common situation.With resource information superiority, some resource service providers (RSPs, sellers) often make the poolingequilibrium, so that resource service demanders (RSDs, buyers) cannot recognise high-quality resources owing toimperfect information Hence, the authors propose a signalling games approach to prevent resource trading fraud inthe manufacturing grid In support of the architecture of resource negotiation and trading, RSDs can get moreaccurate information about resource quality based on the collateral currency promised by RSPs In this way, RSDscan more accurately identify resource quality This method focuses on preventing low-quality RSPs from sendingout an incorrect signal suggesting high resource quality to entice RSDs to purchase the low-quality resource.Simulation results indicate that the game theoretical model has a reasonable and perfect Bayesian separatingequilibrium, from which RSPs do not initiatively deviate

Keywords: manufacturing grid; singling games; the perfect Bayesian equilibrium; trading fraud

1 Introduction

With the rapid development of advanced

manufactur-ing technology, information technology, and the global

market, modern manufacturing enterprises are facing

sustainable, variable, and unpredictable competition

The traditional quality and price competitive model in

manufacturing has turned into a service, technology,

and time competitive model In this situation,

enter-prises cannot continue the traditional mode of ‘big and

all-inclusive projects’; they must adopt a new ‘small

and specialised’ thought processes Modern enterprises

must emphasise professionalism and standardisation

In the new economy, most enterprises adopt a win–win

strategy for cooperation Thus, a common platform

must be developed – one that promotes the sharing of

manufacturing resources and service

The manufacturing grid (MGrid) (Qui et al 2004,

Shi et al 2007) is a new concept and one of the

next generations of manufacturing models It has

been proposed to meet the cooperative demand of

the manufacturing industry It enables geographically

distributed manufacturing resources to be connected

through the internet using grid technology With

the MGrid platform, common resources and services

(including human, equipment, material, and software)

can be shared Enterprise competitiveness can be

enhanced in the MGrid owing to the shortened

development and manufacturing time and minimisedcost The MGrid tries to achieve the sharing ofgeographically distributed manufacturing resourcesand services through the reconfigurable manufacturingprocesses (also called ‘virtual organisations’) There-fore, enterprises in the MGrid can achieve thegoal of TQCSEF (Time, highest Quality, lowestCost, best Service, cleanest Environment, and greatestFlexibility)

With the technical support of Web services(Mockford 2004) and Grid (Foster et al 2001), MGridhas constructed a huge manufacturing-oriented re-source and service market Enterprises and evenindividuals can acquire manufacturing resources andservices from the MGrid resource market as conve-niently as they obtain water, electricity, and internetinformation MGrid users can easily pay for resources

or services (such as the material, design, or machinerythey consume) with the records and payment functionsprovided by the MGrid core middleware

Participants in the MGrid market are divided intotwo categories: they are either resource service pro-viders (RSPs) or resource service demanders (RSDs).Resource quality is the basis of both the RSD’spayment and the RSP’s pricing The aim of bothRSPs and RSDs is to maximise their own profits As

in the commodity market, the resource informationasymmetry is for RSPs and RSDs in the MGrid

*Corresponding author Email: haijun@whut.edu.cn

Vol 23, No 5, May 2010, 391–401

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

Ó 2010 Taylor & Francis

DOI: 10.1080/09511921003642113

resource market With resource information

super-iority, some RSPs send out a confusing signal (or

inaccurate information) regarding resource quality,

which may entice RSDs to buy or use the low-quality

resource to make the pooling equilibrium When this

happens, RSDs cannot recognise a high-quality

resource due to imperfect information, and RSPs

make an abnormal profit This paper focuses on

preventing RSPs who have low-quality resources

from sending out untrue information Therefore,

resource information asymmetry induces two

impor-tant issues in the MGrid resource trading: (1) how to

display resource quality information for RSPs in order

to attract the attention of RSDs and (2) how RSDs can

obtain accurate information about the resource quality

from RSPs

MGrid resource trading involves the individual

behaviours of RSPs and RSDs In free market

economics, whether RSPs provide high-quality

re-sources or low-quality rere-sources is a result of the

rational behaviours of both RSPs and RSDs The

authors adopt the signalling game theory in the

infor-mation economics to analyse individual behaviours

of both RSPs and RSDs The signalling game theory

was chosen because it focuses on decision making and

the related market equilibrium given the interaction

between decision makers

The rest of the paper is organised as follows:

Section 2 investigates the related work about MGrid

Section 3 describes the MGrid resource negotiation

and trading model based on perfect Bayesian Nash

equilibrium Section 4 simulates the model and

analyses the results Section 5 shows an industrial

case for the model The conclusion and future work are

given in Section 6

2 Related work

The conception, system platform, Open Grid Services

Architecture (OGSA) and Web Services Resource

Framework (WSRF) of MGrid have been investigated

by several authors (Li et al 2007, Tao et al 2008a,

Zhang et al 2008) The application demands of grid

technology in e-science, e-government,

e-entertain-ment, e-education, e-business, and manufacturing

industries have also been studied In such MGrid

environments, RSPs and RSDs have different goals,

objectives, strategies, and supply-and-demand

pat-terns Moreover, both resources and end-users are

geographically distributed with different time zones

The economic approach provides a fair basis in

successfully managing the decentralisation and

hetero-geneity that is present in human economies In an

economic-based approach, the resource scheduling is

made dynamically at runtime and is driven and

directed by the end-users’ requirements Pricingbased on user demand and resource supply is themain driver in the competitive, economic MGridresource market

Since MGrid uses the internet as a carrier forproviding remote manufacturing services, MGridcan be used to share manufacturing resources in aseamless manner for cooperative problem solving.The resource price is adjusted, according to the law

of supply and demand Price fluctuation reflectsthe market’s supply-and-demand dynamics, and theoptimal resource allocation occurs at the supply-and-demand equilibrium

The application of economic theory in computersystem resource allocation can be traced back to 1968,when Sutherland proposed the auction mechanism forresource allocation in the PDP-1 computer (Sutherland

et al 1968) Since then, the economic theory hasbeen primarily used for solving the load balance ofcomputer clusters and distributed systems Ferguson(1996) investigated the application of general equili-brium theory and Nash equilibrium in the distributedcomputer resource management Waldspurger (1992)designed and developed the Spawn system, which isthe market-oriented scheduling system for a group ofheterogeneous computers connected to the internet.Bogan (1994) used the market mechanism to allocatethe central processing unit (CPU) time and proposedthe CPU leasing agreement per unit of time Bredin(1998) proposed a trusted third-party arbiter toprevent fraud in transactions between mobile agents,which is a little similar to the notary public in Section3.3 However, this paper focuses on how to let thecollateral currency represent the quality level ofresources and/or services truly

In recent years, studies related to the application ofeconomic theory in grid resource allocation havebecome very popular (Buyya et al 2002, Subramoniam

et al 2002) Abramson (2002) studied the resourcemanagement, scheduling, and computational econom-ics in a grid environment and designed the gridarchitecture for computational economy (GRACE).Buyya (2009) later developed a grid application soft-ware toolkit based on Globus–Nimrod/G Wolski(2000, 2001) proposed that the relative value of aresource varies with the supply-and-demand andallocated dynamic resources using an auction model

in G-commerce project The Popcorn project focused

on formulating computer service time as currency(London 1998)

Computing grid researchers have carried out an depth study on the application of economic theory in alarge-scale, dynamic system–Grid They achieve prettygood results that now provide the references forMGrid

in-More and more researchers are becoming aware of

the importance of economics in MGrid Wu et al

(2005) presented a trading support manufacturing

resource sharing model similar to P2P and deployed

core services at every node responsible for aggregating

manufacturing resource information Giovanni (2002)

investigated the economic convenience in front of

dedicated manufacturing systems depending on the

competitive market conditions They proposed a

theoretical model that can locate market conditions

and make scope economy manufacturing systems less

profitable than dedicated manufacturing ones, and set

some general criteria to guide the entrepreneur in

making wise investment decisions regarding these

kinds of manufacturing investments Lu et al (2007)

defined the competitive equilibrium of Pareto optimal

in the MGrid resource optimal allocation, in order

to realise profit maximisation for RSPs and efficiency

maximisation for RSDs

However, insufficient research has been done on

how to display resource quality information for RSDs

and how to obtain accurate information regarding

resource quality from RSPs There are two traditional

ways of solving this issue First is the centralised trust

system (Resnick and Zeckhauser 2002), which assumes

that a few of the entities record the history of entities

that participate in the network, and calculate and

publish the results of the creditability evaluation for

every entity Second is the distributed trust system

(Cornelli et al 2002), which assumes that every entity

calculates the credibility of a certain entity in the

network, according to the behaviour evaluations

provided by others Information is generally static in

the centralised trust system, while network

commu-nication in the distributed system makes it difficult to

calculate credibility

The MGrid environment has the following

char-acteristics: resource distribution and sharing,

self-similitude, dynamics and diversity, autonomist and

multiplicity of management, and highly abstract and

transparency (Tao et al 2009) The resource

manage-ment and cooperation in MGrid is more complex and

difficult than other types of distributed

informa-tion systems Therefore, the static informainforma-tion in the

centralised trust system and the complex

communica-tion in the distributed system are not suitable for the

MGrid environment It needs a simple and flexible

approach for solving the above-mentioned issues for its

resource market

The resource trading process in MGrid is

essen-tially a game in which RSPs and RSDs evaluate each

other’s behaviour and characteristics, and then choose

the strategies that maximise their own individual

benefits Current state-of-the-art grid technology uses

game theory for grid resource optimised allocation

(Riky et al 2008, Tao et al 2008b) The authorsintroduce the game theory into the resource tradingprocess in MGrid and propose an MGrid resourcetrading model based on the perfect Bayesian Nashequilibrium In this way, the authors analyse thetrading process using a dynamic model of incompleteinformation This model makes RSPs tend not tocheat and RSDs determine the quality of the resourceaccording to the collateral currency promised byRSPs This is a new method of solving the aboveissues of MGrid resource trading The key is to preventRSPs from sending out an improperly high qualitysignal

3 MGrid resource negotiation and trading model3.1 The model of MGrid resource marketConsider resource trading in the MGrid environment:

If RSDs need some manufacturing resource or service,they search for resources and inquire about prices inthe MGrid market, according to the catalogue ofresource RSPs can also register and maintain resourceand service information through the MGrid portal.The information in the MGrid market is uniformlystored and managed by the MGrid index informationserver (MGIIS), as shown in Figure 1 Suppose RSPswho provide the same type of resources but at differentlevels of quality, all charge the same price P The RSPsknow the level of quality of resources they offer (butthe RSDs do not), and they promise a collateralcurrency F If a RSD argues that the purchasedresource does not meet contract specifications, then theRSP compensates the RSD by paying F Therefore,RSDs would like to purchase resources with higher F,given a constant P

The goal of this paper is to establish an effectivesignalling game model in which RSPs are motivated tomake a promise of a collateral currency F thataccurately reflects the quality of their resource Insignalling game theory, the quality of resource can beregarded as the types of RSPs given by ‘nature’, andthe collateral currency F as the signalling according totheir types RSDs infer the quality of a given resourcefrom F and choose one RSP as their supplier Then,the dynamic signalling game model of incompleteinformation would be established in order to solve theabove-mentioned issues in MGrid resource trading

3.2 Signalling gamesSignalling games (Myerson 1997) are dynamic games

in which information transfer is viewed as the signal.They are incomplete information games with twoplayers, a sender and a receiver The type of sender (i.e.the quality of resources that RSPs provide) is private

information; the type of receiver (i.e RSDs prefer to

buy resources with a higher collateral currency, given a

constant price) is public information The two players

receive payoffs depending on the type of sender, the

message chosen by the sender, and the action chosen

by the receiver The sequence of signalling games is as

follows:

(1) The sender has a certain type y2 Y, which is

given by nature Y¼ {y1, , yK} is the set of

sender types The sender knows its own y while

the receiver does not know the type of the

sender The receiver only knows the prior

probability p¼ p(y) where the type of the

sender is y, Skp(yk)¼ 1

(2) Based on his knowledge of his own type, the

sender chooses to send a message (m) from a set

of possible messages M¼ {m1, , mJ}

(3) The receiver observes the message (m) but not

the type of the sender (y), and gets the posterior

probability ~p ¼ ~p (y j m) from the prior

prob-ability p¼ p (y) according to the Bayes rule

Then, the receiver chooses an action a2 A

from a set of feasible actions A¼ {a1, , aH}

(4) The payoff functions of the sender and the

receiver are u1(m, a, y) and u2(m, a, y),

respectively

The equilibrium concept that is relevant for

signalling games is Perfect Bayesian equilibrium

Perfect Bayesian equilibrium is a refinement of

Bayesian Nash equilibrium, which is an extension of

Nash equilibrium of incomplete information games

Perfect Bayesian equilibrium is the equilibrium

concept relevant for dynamic games of incompleteinformation

3.3 Architecture of MGrid resource negotiation andtrading

For MGrid resource trading, an architecture forMGrid resource negotiation and trading is designed

in the section The architecture is sufficiently general toaccommodate different economic models used forresource trading and service access cost determination(see Figure 2 for more details) This consists of fourparts: RSDs, grid brokers, core grid middleware, andRSPs The features of GRACE (Abramson et al 2002)have been extended to focus on preventing resourcetrading fraud (such as the credit management module)

On the basis of the resource negotiation andtrading architecture, the resource trading process is

as follows (shown in Figure 3):

(1) RSPs and RSDs login to MGIIS

(2) RSPs and RSDs deposit money into the GridBank In this study, the authors use a ‘prepaid’payment mechanism

(3) RSPs publish their resource information (i.e.process capability, price, duration, and collat-eral currency) in the MGrid market through theMGrid portal to attract RSDs RSDs can alsosearch and check resource information throughthe portal

(4) RSPs sign a contract with RSDs In themeantime, the contract must be legally no-tarised and the notary public takes over bothdeposits in the Grid Bank

Figure 1 The MGrid resource market model

Figure 2 The architecture of resource negotiation and trading in MGrid.

(5) If a trade is successful, the RSD’s deposit will

be transferred into the RSPs account by the

notary public Otherwise, the RSP’s deposit is

transferred into the RSD’s account as

compen-sation No matter what, the credit system

records this trading information about RSDs

and RSPs for later credit inquiry

3.4 Model analysis

According to Taguchi et al (2005), quality implies low

failure rate, low energy consumption, long service life,

high efficiency, and low damage to users For

convenience, assume the quality level of the resource

is q (0 q  1), where higher q indicates higher

resource quality and q¼ 1 means that the resource

has a zero failure rate The authors employ the

ex-ponential distribution for the failure rate G(q) ¼ e–kq,

where k 2 Rþsuch that k 4 0 RSPs charge P for

each resource and promise a collateral currency F The

cost of a resource is given by C(q) ¼ a1e–b1q þ a2e–b2q

(Juran et al 1999), where a1 refers to the loss of

defective resource as q! 0, a2refers to the cost of the

resource as q! 1, and b1, b2are the coefficients of the

function From a practical perspective, P  C(q) 4 0,

F4 0 The action of RSDs is to determine the trading

probability Prob() In general, RSDs consider two

factors: quality and price (that is, the cost

perfor-mance) At the same P, the higher F that RSPs

promise, the higher qualities of resource RSDs assume

Therefore, RSDs prefer to trade with RSPs who have

the highest F, such that dProbðFÞdF >0 Therefore, theformulae of Prob() is given by

ProbðFÞ ¼ meq Fð Þ

where 0 5 Prob(F) 1, m 4 0,0 5 n  1, m and nare the adjustment coefficients ~q(F) is the estimation ofresource quality after the RSD observed compensationprice F

The MGrid resource trading model based onperfect Bayesian Nash equilibrium is described asfollows:

(I) ‘Nature’ chooses one resource quality qaccording to a certain prior probabilitydensity, and informed RSPs know it as theirtypes

(II) RSPs fix F

(III) RSDs observe F without knowledge of q, andthen determine the trading probabilityProb(F)

(IV) The payoff function of RSPs is U (q, F, Prob(F))

Define the benefit as the added value of profit, notconsidering the funding rate If there is no trading, thebenefit is zero Suppose that RSPs are risk-neutral;therefore, the expected benefit of RSPs is

Uðq; F; ProbðFÞÞ ¼ P  F  C q½ð ð ÞÞ  G qð Þ

þ P  C qð ð ÞÞ  1  G qð ð ÞÞ  ProbðFÞ

Figure 3 The flow chart of resource trading in MGrid.

Simplify it and get the following function:

Uðq; F; ProbðFÞÞ ¼ P  F  G q½ ð Þ  C qð Þ  ProbðFÞ

Furthermore, the price F should satisfy U(q,F,

Prob(F)) 0, so the constraint of F can be obtained as

follows:

FP CðqÞ

3.5 Model solution

As a pure strategies equilibrium, the above model may

have the solutions of a pooling equilibrium, either

separating equilibrium or semi-separating equilibrium

The goal of this study is to get the solution of

separating equilibrium, that is, the higher quality ofresource, the more likely RSPs are to promise a highercompensation price

Based on perfect Bayesian Nash equilibrium,RSPs and RSDs plan optimal reactions in (II) and(III), against the possible strategies of their oppo-nents for each potential type of their own Thereaction of a RSP depends on their type Hence,

an RSP’s reaction reveals some information abouttheir type RSDs can infer their opponent’s type

or revise the prior probability, and then choosethe optimal reaction RSPs know that their reac-tions will be known or utilised by RSDs, and there-fore try to choose reactions that most benefitthemselves

Suppose that there exists the partial derivative of U

to F, and let@U¼ 0,

If RSPs promise the collateral currency F for their

resource, u˜(qj F) is the posterior probability of q after

RSDs observe the price F Then the expected value of

the resource quality for RSDs isqðFÞ ¼e R1

0 q~uðq j FÞdq

u˜(qj F) can be calculated according to the Bayes

formula – a theorem that is valid in all common

interpretations of probability – in the following way:

~ðq j FÞ ¼R1PðF j qÞuðqÞ

0 PðF j qÞuðqÞdq

Regarding the information set F observed by

RSDs, the belief of RSDs 7 u˜(qj F) satisfies

R1

0uðq j FÞdq ¼ 1.e

For separating equilibrium, RSDs can infer q from

F because F(q) is the optimal reaction of RSPs who

provide a resource of quality q So there is

deq F qð ð ÞÞ=dq ¼ 1

In fact, there are P(qj F(q)) ¼ 1and P(q0 j F(q))

0 for the separating equilibrium,where q0 6¼ q

By combining Equations (1), (3), and (4), the

following formula can be obtained:

By combining C(q) and G(q) with the above

formula and solving the differential equation, the

expected value F is obtained

Let ~q(F)¼ q ¼ y–1(F*), under the observation of

rational expectations (two parties can identify the type

accurately), the payoff function of RSPs is

do not consider there are resources with the quality

q 5 q0 In Figure 4, F1represents the maximum of thecollateral currency which satisfies the inequality (2); F2represents the optimal value of the compensate pricethat satisfies Equation (3) RSPs with resource quality

q can gain the maximum profit by promising acompensation price F2 The curve L2 in Figure 4implies that F* should range from F3to F4 Therefore,prob(F)¼ 0 when F 5 F3, while Prob(F)¼ 1 when

F4 F4 Therefore, the trading probability is given by

(3) For convenience of analysis, the authorsassume the quantity of the trading resource isone in the model However, the model is thesame for other quantities

Figure 4 The compensation price & resource qualitycurves

(4) The model takes into account the cost of a

resource, given different actual qualities, while

the literature (Li et al 2005) does not Because

the different quality resources have different

costs, obviously, the profit is different at the

same price

4 Model simulation

The authors employ MATLAB V7.6.0 as the

simula-tion tool and personal computer (Intel Pentium 4 CPU

3.40 GHz and 2GB memory, operation system:

Win-dows Vista Enterprise) as the simulation platform

4.1 Simulation parameter set

A machine plant (RSD) wants to purchase a certain

part, for which all machine tool plants (RSPs) charge

P¼ 13.50 However, the quality of the parts provided

by different machine tool plants may vary So, the

machine plant plays games with machine tool plants

In the MGrid resource market, machine tool plants

promise different compensation prices based on

the quality of their part The other parameters are

C1¼ 1.00, m ¼ 13.095; n ¼ 0.03 According to the

historical data of products, a1¼ 2.64, b1¼ 0.98,

a2¼ 0.31, b2¼ 3.35, k ¼ 5.00 can be obtained by

means of statistical computing The cost of a part is

given by C(q) ¼ 2.64e–0.98qþ 0.31e3.35qand the failure

rate is given by G(q)¼ e–5q According to the above

resource trading model, the expected profit

distribu-tion for the different resource qualities and the

different compensation prices is shown in Figure 5

In order to validate the model, q is set to be 0.38,

0.68, and 0.98, respectively Figure 6 shows the 36

expected profits of machine tool plants at the collateralcurrency F for every two dollars on the closed interval[0,F1], when the quality of resource provided by themachine tool plant is 0.38 The machine tool plants canobtain the maximum profit (Umax¼ 2.15) at thecollateral currency F2¼ 34.59, while q is 0.38 Figure

7 shows the 28 expected profits at the collateralcurrency F for every ten dollars on the closed interval[0,F1], when q is 0.68 The machine tool plants achievethe maximum profit (Umax ¼ 4.35) at the collateralcurrency F2¼ 84.82 for q ¼ 0.68 Figure 8 shows the

29 expected profits at the collateral currency F forevery ten dollars on the closed interval [0,F1] for

q ¼ 0.98 The machine tool plants achieve the imum profit (Umax¼ 2.73) at the collateral currency

max-F2¼ 193.89 for q ¼ 0.98

Figure 5 The expected profit distribution for RSPs

Figure 6 The expected profit distribution of RSPs at

q¼ 0.38

4.2 Analysis of simulation results

(1) When the collateral currency F promised by a

RSP is the optimal value F2corresponding to q,

the RSP achieves the maximal profit

(2) The optimum value F2 increases with the

resource quality, and RSPs always pursue their

own individual profit maximisation in

signal-ling games RSDs should be able to infer the

resource quality q from F correctly

(3) When F is greater than F1or less than F3, the

expected profit is less than zero

These results are feasible No one would like topurchase a resource that has a very low collateralcurrency Conversely, RSPs may take risks by payingRSDs more collateral currency in the case of resourcebreakdown if the collateral currency is too high.The simulation results indicate that the gametheoretical model has a reasonable and perfect Bayesianseparating equilibrium, from which RSPs would notinitiatively deviate This means that in pursuit ofmaximal profits, RSPs prefer to provide RSDs withaccurate information about their resource’s quality (i.e.the compensation price) initially and faithfully, giventhe strategy of RSDs (i.e the trading probability) in the

Figure 7 The expected profit distribution of RSPs at

signalling games model If RSPs do not choose the

optimum value F2as their collateral currency, then they

do not achieve the maximal profit in resource trading

Therefore, RSPs prefer to choose a collateral currency

according their actual resource quality as opposed to

choosing an incorrect collateral currency that is higher

or lower than F2 hoping to cheat RSDs Then, RSDs

can identify the resource quality according to the

collateral currency F2 because a higher F2 implies a

higher q

5 Industrial case study

This methodology is employed in the development

and production of magnetic bearing Magnetic

bear-ing is a new type of high–performance bearbear-ing

Compared with ordinary bearings, magnetic bearings

have lots of unique advantages, such as no

mechan-ical contact, no friction, no wearing, no lubrication,

and a long lifetime However, each type of magnetic

bearing must be designed and manufactured,

accord-ing to the concrete applications A large number of

resources are needed in the development of magnetic

bearings

In order to realise the sharing and collaboration

of resources, the magnetic bearing resource and

service sharing platform in MGrid (MBRSSP-Grid)

was developed The feasibility for the implementation

of the model is demonstrated in this experimental

platform MBRSSP-MGrid is implemented on

Glo-bus Toolkit 4.0 Resources and services are

encapsu-lated in the form of Web Services Resource

Framework (WSRF) RSPs can publish the

informa-tion of their valued resources and services (such as

equipment, software, human, application, and

tech-nique) through the publication system of

MBRSSP-MGrid Users can also search and schedule resources

or services by inquiring through the resource and

service optimal-selection (RSOS) system of

MBRSSP-MGrid Figure 9 is a screenshot of the man-machine

interface of the RSOS system RSOS provides users

with the list of services sorted by the highest collateral

currency first

6 Conclusions and future work

In support of the architecture of resource negotiation

and trading, the authors propose an MGrid resource

trading model based on perfect Bayesian Nash

equilibrium in order to prevent trading fraud The

model is essentially a dynamic game of incomplete

information, which best conforms to the realities of

resource trading The theoretical analysis and

simula-tion results indicate that the model is effective in

preventing low-quality RSPs from sending out

incorrect signals regarding their resource quality(which may entice RSDs to buy or use the low-qualityresources) Since through this model RSDs obtainaccurate information about resources, a recommendedfuture work includes the combinatorial optimisation ofresource service in the MGrid, given one or morecriterion, such as minimised time, minimised cost, theresource utilisation, and so forth

Acknowledgements

This paper is supported by the National Natural ScienceFoundation Key Project of China: Digit manufacturing basictheories and key techniques under network environment(NO.50335020), and the Hubei Digital Manufacturing KeyLaboratory Opening Fund project: Research on resourceservice search and optimal-selection theories and experiments

in manufacturing grid system (No SZ0621) The authorsthank the editor and the anonymous reviewers for theirconstructive comments and suggestions which helped toimprove the paper

References

Abramson, D., Buyya, R., and Giddy, J., 2002 A tional economy for grid computing and its implementa-tion in the Nimrod-G resource broker Future GenerationComputer Systems, 18 (8), 1061–1074

computa-Bredin, J., Kotz, D., and Rus, D., 1998 Market-basedresource control for mobile agents In: 2nd InternationalConference on Autonomous Agents, 9–13 May, Minnea-polis, MN New York: ACM Press, 197–204

Bogan, N.R., 1994 Economic allocation of computation timewith computation markets Thesis (MS) MassachusettsInstitute of Technology

Buyya, R., Abramson, D., Giddy, J., and Stockinger, H.,

2002 Economic models for resource management andscheduling in Grid computing Concurrency and compu-tation: Practice and Experience, 14 (13–15), 1507–1542.Buyya, R., Abramson, D., and Giddy, J., 2009 The Nimrod/

G grid resource broker for economic- based scheduling,market oriented grid and utility computing New Jersey:Wiley Press

Cornelli, F., Damiani, E., Vimercati, S.D., Paraboschi, S.,and Sanmarati, P., 2002 Choosing reputable servants

in a P2P network In: D Lassner, ed 11th InternationalWWW Conference, 7–11 May, Hawaii, USA New York:ACM Press, 376–386

Ferguson, D.F., Nikolaou, C., Sairamesh, J., and Yemini,Y., 1996 Economic models for allocating resources incomputer systems Market-based Control: A Paradigmfor Distributed Resource Allocation, 156–183

Foster, I., Kesselman, C., and Tuecke, S., 2001 The anatomy

of the grid: enabling scalable virtual organizations.International Journal of High Performance ComputingApplications, 15 (3), 200–222

Giovanni, P., Michele, A., Giovanna, L.N., and Sergio,N.L.D., 2002 Long term capacity decisions in uncertainmarkets for advanced manufacturing systems incorpor-ating scope economies European Journal of OperationalResearch, 143 (1), 125–137

Juran, J.M., Godfrey, A.B., Hoogstoel, R.E., and Schilling,E.G., 1999 Juran’s quality handbook 5th ed New York:McGraw Hill

Li, M.S., Yang, S.B., Fu, Q.F., and Jin, Y., 2005 Research

on grid resource reliability model based on promise In:

International Conference on Information Technology:

Coding and Computing, 4–6 April 2005, Las Vegas, NV

New Jersey: Institute of Electrical and Electronics

Engineers Inc., 310–339

Li, Z., Jin, X.L., Cao, Y., Zhang, X.Y., and Li, Y.Y., 2007

Conception and implementation of a collaborative

manufacturing grid International Journal of Advanced

Manufacturing Technology, 34 (11–12), 1224–1235

London, S., 1998 POPCORN – a paradigm for

global-computing Thesis (M.S.) The Hebrew University of

Jerusalem

Lu, S., Zhou, Y.L., Cao, X.L., and Liu, T., 2007 Study on

economic architecture for resource management for MG

based on beneficial drive In: International Conference on

Wireless Communications, Networking and Mobile

Com-puting, 21–25 September 2007, Shanghai, China New

Jersey: Institute of Electrical and Electronics Engineers

Inc., 3411–3414

Mockford, K., 2004 Web services architecture BT

Technol-ogy Journal, 22 (1), 19–26

Myerson, R.B., 1997 Game theory: analysis of conflict

Cambridge, MS: Harvard University Press

Qiu, R.G., 2004 Manufacturing grid: a next generation

manufacturing model In: IEEE International Conference

on Systems, Man and Cybernetics, 10–13 October, The

Hague, Netherlands New Jersey: Institute of Electrical

and Electronics Engineers Inc., 4667–4672

Resnick, P and Zeckhauser, R., 2002 Trust among strangers

in internet transactions: empirical analysis of eBay’s

reputation system In: M.R Baye, ed The economics of

the internet and e-commerce Bingley, UK: Emerald

Group Pub Ltd., 127–157

Riky, S., Zomaya, A.Y., and Bjorn, L., 2008 A cooperative

game framework for QoS guided job allocation schemes

in grids IEEE Transactions on Computers, 57 (10), 1413–

1422

Shi, S.Y., et al., 2007 An implementation of modelling

resource in a manufacturing grid for resource sharing

International Journal of Computer Integrated

Manufac-turing, 20 (2–3), 169–177

Subramoniam, K., Maheswaran, M., and Toulouse, M.,

2002 Towards a micro-economic resource allocation in

grid computing systems In: IEEE conference on electrical

and computer engineering, 12–15 May, Winnipeg,

Man-itoba, Canada, New Jersey: Institute of Electrical and

Electronics Engineers Inc., 782–785

Sutherland, E., 1968 A futures market in computer time.Communications of the ACM, 11 (6), 449–451

Tao, F., Hu, Y.F., and Zhou, Z.D., 2008a Study onmanufacturing grid and its executing platform Interna-tional Journal of Manufacturing Technology and Manage-ment, 14 (1–2), 35–51

Tao, F., Hu, Y.F., and Zhou, Z.D., 2008b Resource servicecomposition and its optimal-selection based on particleswarm optimization in manufacturing grid system IEEETransactions on Industrial Informatics, 4 (4), 315–327.Tao, F., Hu, Y.F., and Zhou, Z.D., 2009 Application andmodeling of resource service trust-QoS evaluation inmanufacturing grid system International Journal ofProduction Research, 47 (6), 1521–1550

Taguchi, G., Chowdhury, S., and Wu, Y., 2005 Taguchi’squality engineering handbook New Jersey: Wiley Press.Waldspurger, C.A., Hogg, T., Huberman, B.A., Kephart,J.O., and Stornetta, W.S., 1992 Spawn: A distributedcomputational economy IEEE Transactions on SoftwareEngineering, 18 (2), 103–117

Wolski, R., Plank, J.S., Brevik, J., and Bryan, T., 2000.Analyzing market-based resource allocation strategiesfor the computational grid International Journal of HighPerformance Computing Applications, 15 (3), 258–281.Wolski, R., Plank, J.S., Brevik, J., and Bryan, T., 2001 G-commerce: market formulations controlling resourceallocation on the computational In: 15th internationalparallel and distributed processing symposium, 23–27April, San Francisco, CA New Jersey: Institute ofElectrical and Electronics Engineers Inc., 262–269

Wu, L., Meng, X.X., Liu, S.J., Tong, Y.X., and Wei, M.,

2005 A trading supported manufacturing resourcesharing model for manufacturing grid In: 9th interna-tional conference on computer supported cooperative work

in design, 24–26 May, Coventry, United Kingdom, NewJersey: Institute of Electrical and Electronics EngineersInc., 339–344

Zhang, H.J., Hu, Y.F., and Zhou, Z.D., 2008 Study onsemantic-aware manufacturing grid architecture In: 5thinternational conference on fuzzy systems and knowledgedsicovery, 18–20 October, Shangdong, China NewJersey: Institute of Electrical and Electronics EngineersInc., 626–630

Integrated line balancing to attain Shojinka in a multiple straight line facility

Hadi Go¨kc¸ena, Yakup Karab* and Yakup Atasagunba

Department of Industrial Engineering, Faculty of Engineering and Architecture, Gazi University, Maltepe, 06570, Ankara, Turkey;

b

Department of Industrial Engineering, Faculty of Engineering and Architecture, Selc¸uk University, Campus, 42031, Konya, Turkey

(Received 3 January 2009; final version received 14 January 2010)

Traditional straight assembly lines are still one of the most important elements and an important fact of today’sproduction systems If applicable, a company can combine its multiple straight assembly lines and obtain manyadvantages of Shojinka more or less This paper analyses a new problem – integrated balancing of multiple straightassembly lines (MSLB) to attain Shojinka in a multiple straight assembly line facility The MSLB problem is built onthe concept that it could be possible for a company to obtain the advantages of Shojinka even if the company hasnot adopted the U-shaped line layout Three connectivity types are suggested to integrate multiple assembly lines

A binary integer formulation for integrated balancing of multiple assembly lines is developed The objective of theproposed formulation is to minimise the total number of workstations required in the assembly facility Theformulation is explained and validated using some illustrative examples The proposed approach provides flexibility

to minimise the total idle times of the lines and total number of workstations that are required in the assembly linefacility

Keywords: assembly line balancing; integer programming; Shojinka

1 Introduction

‘Shojinka’ is a Japanese word that is a combination of

sho (to reduce), jin (worker) and ka (to change)

(Sennott et al 2006) The concept of Shojinka, which

was originally an important element of Toyota

Production System (TPS), is easily to increase or

decrease the number of workers in a production facility

when the demand rate is increased or decreased In a

production facility, different types of products may be

produced on different lines The fluctuations in

demands of products will probably require adding

workers to some lines and removing from others

Attaining flexibility in the number of workers at a

workshop to adapt to demand changes is called

Shojinka, which is equivalent to increasing

productiv-ity by the adjusting and rescheduling of human

resources (Monden 1993) Shojinka can be attained

by changing the number of operations assigned to a

worker, who is capable of performing multiple

opera-tions On the other hand, the machinery layout must

be proper to enable workers easily to walk between

machines The TPS adopts U-shaped machinery layout

to form production lines The U-shaped layout has

several advantages over other layout types such as bird

cages, isolated islands and linear (straight) layouts

(Monden 1993) The visibility and communication

between workers on U-shaped production lines are

strengthened In addition, the number of workersrequired on a U-shaped line will be less than or equal

to the number of workers required on a comparablestraight line (Miltenburg and Wijngaard 1994) For agiven cycle time, the number of workers required on aU-shaped line may be fractional such as 3.4 workers.Four workers must be assigned to this U-shaped line tomeet the demand Hence, 60% of available time forone of the workers will be idle (waste) on this line That

is, (4.0–3.4)/4.0¼ 15% of total available workforce

is unproductive In order to make this idle timeproductive, several U-shaped lines are combined intoone integrated line by locating them close to eachother This way, a worker can perform operationsfrom two or more neighbour U-shaped lines, and idletimes can be eliminated or reduced Such a productionfacility that consists of a group of U-shaped lines

is defined as a Just-in-Time (JIT) production unit(Sparling 1998)

The problem of assigning operations to stations on U-shaped lines, the single U-line balanc-ing (SULB), was first studied by Miltenburg andWijngaard (1994) Essentially, the SULB problem isthe U-shaped line version of a well-known problem,the simple assembly line balancing (SALB), which wasintroduced by Salveson (1955) Although numerousresearch papers on SALB have been published, the

work-*Corresponding author Email: ykara@selcuk.edu.tr

Vol 23, No 5, May 2010, 402–411

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

Ó 2010 Taylor & Francis

DOI: 10.1080/09511921003642162

literature on SULB is relatively small It is not

practical to present the literature on SALB here

Nevertheless, several review and assessment papers can

be useful for interested readers (Baybars 1986, Ghosh

and Gagnon 1989, Erel and Sarin 1998, Becker and

Scholl 2006) Following the pioneering study of

Miltenburg and Wijngaard (1994), many researchers

attempted to develop SULB procedures (Urban 1998,

Scholl and Klein 1999, Erel et al 2001, Aase et al

2004, Go¨kc¸en and Agpak 2006) Since the U-shaped

lines are a consequence of Shojinka, their advantages

can be exactly obtained when they are combined in

comparison to their independent cases Balancing

combined multiple U-shaped lines (or JIT production

units) was studied by only Sparling (1998), Miltenburg

(1998) and Chiang et al (2007) These studies refer to

the same problem by different names as the N U-line

balancing problem with travel (NULB-T) by Sparling

(1998), the U-line facility problem (ULF) by

Milten-burg (1998) and multiple U-line balancing problem

(MULB) by Chiang et al (2007)

The U-shaped production lines mentioned and

emphasised by Monden (1993), Sparling (1998),

Miltenburg (1998) and Chiang et al (2007) mostly

consist of manufacturing operations which need

machinery But, this does not mean that the U-shaped

lines cannot be used for assembly operations

Further-more, most SULB studies emphasise assembly

opera-tions On the other hand, it could be possible for a

company to obtain the advantages of Shojinka even if

the company has not adopted the U-shaped line

layout It should be noted that traditional straight

assembly lines are still one of the most important

elements and an important facet of today’s production

systems Therefore, if applicable, a company can

combine its multiple straight assembly lines and obtain

many advantages of Shojinka more or less In a

multiple assembly line facility, the parent-component

relationships between the items of a final product

designate the location of multiple assembly lines

Depending on the structure of the final product and

production policies, a company may have one or more

sub-assembly line In many industries such as

auto-motive, aircraft, electronics, machinery, etc.,

compa-nies usually assemble subassemblies and final products

in their own production systems and purchase parts

and components from outside vendors If multiple

assembly lines are close to each other then they can be

balanced simultaneously using some connectivity

con-ditions between these lines Balancing multiple

assem-bly lines simultaneously is originally an integration of

assembly lines This requires integrated balancing of

multiple lines and provides an opportunity to reduce

the number of workstations utilised in such production

systems

The literature on balancing straight assembly linesincludes examples of combining multiple straightassembly lines Go¨kc¸en et al (2006) suggested that in

a production facility, two or more straight assemblylines can be located in parallel and they can bebalanced simultaneously They developed a binaryformulation and proposed a heuristic procedure forbalancing of parallel straight assembly lines with theobjective of minimising the number of workstationsrequired in the system Balancing parallel assemblylines simultaneously will constitute workstations con-taining tasks on two adjacent and parallel lines andthese common (multi-line) workstations will providethe flexibility to minimise the total idle times of thelines and total number of workstations required inthe production facility Go¨kc¸en et al.’s (2006) studyassumes that multiple assembly lines are located inparallel and parallel connections can be constructedbetween two lines Essentially, Go¨kc¸en et al.’s (2006)study can be considered as a partial implementation

of Shojinka in a parallel assembly line facility Go¨kc¸en

et al (2006) provide a framework that combines twoparallel assembly lines with common workstations.However, in practice, two or more assembly lines can

be combined with common workstations using severalconnectivity opportunities But, there is no study thatcombines straight assembly lines to attain Shojinka in

a multiple straight line facility This is the first studythat proposes an assembly line balancing procedure toattain the benefits of Shojinka in a multiple straightline facility This paper analyses a new problem, which

is the integrated balancing of multiple straightassembly lines (MSLB) The objective of the problem

is to assign tasks to a minimum number of stations on multiple straight assembly lines Possiblelocation forms and connectivity opportunities amongstraight assembly lines are investigated and given inSection 2 Based on the connectivity types detected, abinary formulation for MSLB is presented in Section 3.The objective of the proposed formulation is tominimise the number of workstations required in amultiple straight assembly line facility The proposedformulation is illustrated using examples in Section 4.Some concluding remarks and opportunities forfurther research are presented in Section 5

work-2 Combining multiple straight assembly linesTwo assembly lines can be combined by connectingthem with one or more common workstations Thus,operators can work in two or more different assemblylines at the same time These connections may provide

an opportunity to reduce the total labour requirement

of the system In this section, we propose that multiplestraight assembly lines can be combined using several

connectivity opportunities between these lines Go¨kc¸en

et al (2006) suggested that two or more straight

assembly lines can be located in parallel to each other

and they can be balanced simultaneously Balancing

parallel assembly lines may result in common

work-stations that include tasks from two adjacent lines

These common workstations will provide flexibility to

minimise the total idle times of lines and total number

of workstations required in the production system

Go¨kc¸en et al.’s (2006) approach is based on the

concept of parallel connectivity of two adjacent

assembly lines located in parallel Figure 1 shows two

simultaneously balanced parallel assembly lines with

parallel connectivity

It can be shown in Figure 1 that the parallel

connectivity appears in two common workstations

The first common workstation includes task 3 of line 1

and tasks 2 and 3 of line 2 The second common

workstation includes tasks 4 and 6 of line 1 and task 5

of line 2 Parallel connectivity can be benefited when

two or more assembly lines are located in parallel and

they are sufficiently close to each other However, it

should be noted here that a common workstation

includes tasks from only two adjacent lines This is due

to the difficulty and inefficiency for operators to travel

from one line to another non-adjacent line

In a production system, it may not be always

possible to locate all assembly lines in parallel

Furthermore, if assembly lines are tightly related to

each other and some supplier–customer relationships

exist between these assembly lines then they may be

located in different ways In addition to the parallel

connectivity, we suggest two new connectivity types,

which are consecutive and perpendicular connectivity It

should be noted here that one or more common

workstations are required to obtain a parallel

connectivity between two assembly lines However,only one common workstation can be utilised to obtainconsecutive or perpendicular connectivity between twoassembly lines

The consecutive connectivity can be benefited whentwo assembly lines are located consecutively Thisconnectivity type can appear between an upstream line(supplier) and its downstream line (customer) If theoutput of an upstream line is the main input (part) of adownstream line, then assembly line managers maydesire to locate these lines consecutively and close toeach other This means that the output of the upstreamline is processed throughout most of the tasks on thedownstream line In this case, we will have the chance

to obtain a consecutive connectivity between theselines Figure 2 shows two simultaneously balancedconsecutive assembly lines

As shown in Figure 2, two consecutive assemblylines are connected with a common workstation thatincludes tasks 4 and 5 of line 1, and task 1 of line 2

A consecutive connection must be constructed at theend of the upstream line and the beginning of thedownstream line

The outputs of an upstream line may not always bethe main part for a downstream line In other words,the outputs of the upstream line may not be processedthroughout most of the tasks on the downstream line.The outputs may be components that are attached tothe main parts processed on the downstream line Thisattachment can be performed at a stage (task) ofassembly process In this case, assembly line managersprobably desire to locate the upstream line to thenearest point of use so as to minimise materialhandling from the upstream line to the downstreamline If such a location exists in a production system,two lines can be connected to each other withperpendicular connectivity Figure 3 shows two assem-bly lines that have a perpendicular connectivity with acommon workstation

Figure 3 shows that two lines are connected with acommon workstation which includes task 3 of line 1and task 5 of line 2 The outputs of the upstream lineare required at the point of task 5 of the downstreamline That is, task 5 cannot be completed unless theoutputs of line 1 feed it

Based on the locations, an assembly line of aproduction system can be connected to another

Figure 1 A parallel connectivity

Figure 2 A consecutive connectivity

assembly line with parallel, consecutive or

perpendi-cular connectivity Two assembly lines can be

con-nected to each other with only one of these

connectivity types at the same time However, an

assembly line can be connected to several assembly

lines with different connectivity types at the same time

Such connectivity cases can be called a mixed

connectivity Figure 4 shows a mixed connectivity of

four assembly lines

Figure 4 shows that line 1 and line 2 are connected

to each other with a parallel connectivity; line 2 and

line 3 are connected to each other with a perpendicular

connectivity; line 2 and line 4 are connected to each

other with a consecutive connectivity As shown in

Figure 4, line 2 is connected to three lines with all types

of connectivity at the same time In addition, line 1 is

not allowed to connect to line 3 and line 4

Balancing several assembly lines simultaneouslyusing their connectivity opportunities will provideflexibility to minimise the total idle times of the linesand total number of workstations required in theproduction system However, there is an importantproblem when connecting multiple assembly lines withcommon workstations If the cycle times of assemblylines vary, the cycle time for a common workstation isthe minimum of the cycle times of multiple assemblylines that it spans (Miltenburg 1998) For example,consider a common workstation that spans line a,where the cycle time is 5 min and that spans line b,where the cycle time is 10 min If the workload of thisworkstation exceeds 5 min, the operator will be unable

to complete tasks on line a in its cycle time

3 Mathematical formulationThe most important management problem in assemblylines is the Assembly Line Balancing (ALB) Thesimplest version of ALB problems is the SALBproblem mentioned above ALB is the problem ofassigning assembly tasks to successive workstations bysatisfying some constraints and optimising a perfor-mance measure This performance measure is usuallythe minimisation of the number of workstationsutilised in the assembly line For a given cycle time,minimising the sum of station idle times is equal tominimising the number of opened stations SALBproblems under this objective are called Type-1.Conversely, if the number of stations is given, thenminimising the cycle time guarantees minimum idletimes, which is known as Type-2 If both number ofstations and the cycle time can be altered, the lineefficiency E is used to determine the quality of a

Figure 3 A perpendicular connectivity

Figure 4 A mixed connectivity

balance The corresponding SALB problem was hence

labeled Type-E (Boysen et al 2007) It is usually

assumed in ALB literature that each workstation

requires one worker Hence, if the number of

work-stations utilised in the line is minimised then the

number of workers required is also minimised and

productivity is maximised There are three main

constraints of ALB problems The first set of

constraints is called assignment constraints that ensure

each task is assigned to at least and at most one

workstation Cycle time constraints are the second set

of constraints that guarantee the workload of a

workstation will not exceed the cycle time Cycle time

is the time interval between two successive completed

products and represents the output rate of the line In

other words, an assembly line produces one product

for each cycle time unit Therefore, the workloads

of workstations should not exceed cycle time The

precedence relationships among tasks are satisfied by

precedence constraints The precedence relationships

among assembly tasks are illustrated using precedence

diagrams Figure 5 shows an example precedence

diagram with six tasks

In this section, we propose a binary formulation for

MSLB The proposed formulation attempts to balance

an entire production system which consists of

combined assembly lines The proposed formulation

is aiming at integrated balancing of multiple assemblylines to attain Shojinka in a multiple straight linefacility using the connectivity types mentioned inSection 2 Therefore, we need an integrated precedencediagram that represents all precedence diagrams ofmultiple assembly lines We suggest that the integratedprecedence diagram of a production system can beobtained by merging individual precedence diagramsinto a huge diagram Multiple precedence diagramscan be integrated based on the connectivity opportu-nities between multiple assembly lines For example, if

a consecutive connectivity exists between two assemblylines, the precedence diagram of the downstreamassembly line can be appended at the end of theupstream assembly line Thus, these assembly lines can

be connected with a common workstation whichincludes at least the last task of upstream line andthe first task of downstream line Based on the sameconnectivity opportunities given in Section 2, Figure 6shows the integrated precedence diagram of multipleassembly lines shown in Figure 4

Although the production system has a SALBproblem for each line, these SALB problems are nowtransformed into a single MSLB problem using theintegrated precedence diagram It should be noted inFigure 6 that task numbers are sequentially modified inthe integrated diagram to prevent any inconvenience.Since line 1 and line 2 are located in parallel and theycan be combined with parallel connectivity opportu-nities, no precedence relationships between any tasks

of these lines are established in the integrated diagram

If a supplier–customer relationship exists between anupstream and a downstream assembly line, a

Figure 5 An example precedence diagram

Figure 6 An integrated precedence diagram

precedence relationship between two assembly lines

should be established There is no supplier–customer

relationship between line 1 and line 2 Once the

integrated precedence diagram is established and

connectivity options between assembly lines are

iden-tified, the following formulation can be used for

simultaneous balancing of multiple assembly lines:

Indices:

i,r,s,k,l : task

h,g,a,b,p,q : assembly line

Parameters and sets:

Ih : set of tasks on line h

J : set of workstations; j¼ 1, .,Kmax

A : set of assembly lines

nh : number of tasks on line h

H : total number of assembly lines in the

production system

production system; N¼P

8h2Anh

Kmax : maximum number of workstations

Ch : cycle time of line h

ti : completion time of task i

S : set of all precedence relationships in

the production system(r,s)2 S : a precedence relationship; task r is an

immediate predecessor of task s

R : set of disconnection relationships

(p,q)2R : a disconnection relationship; line p

cannot be connected with line q

F : set of perpendicular connectivity

relationships(a,b)2F : a pair of lines; upstream line a can be

connected to downstream line b withperpendicular connectivity

k : the last task in the precedence diagram

of upstream line a

l : the task of downstream line b to which

the outputs of upstream line a are inputVariables:

xhij : 1, if task i of line h is assigned to

workstation j; 0, otherwise

Uhj : 1, if workstation j is utilised on line h; 0,

otherwise

zj : 1, if workstation j is utilised; 0, otherwise

W(a,b)j : 1, if workstation j includes tasks from

both line a and line b That is, it is a

common workstation; 0, otherwise

V(a,b)j : 1, if workstation j includes tasks from

only line a or line b That is, it is not a

common workstation; 0, otherwise

We assume in our formulation that completiontimes of tasks are deterministic and the longest tasktime is not greater than cycle time The mathematicalformulation is presented below:

Uhj Hzj 0 8j 2 J ð11Þ

The objective of the proposed formulation is tominimise the number of workstations utilised in theentire production system Equation (2) ensures thateach task is assigned to at least and at most oneworkstation Equation (3) ensures that task s cannot beassigned until its predecessor task r is assigned to anearlier or the same workstation that task s is assigned

Equation (4) is the cycle time constraints of the model.

This set of constraints ensures that the work content of

a workstation does not exceed cycle time These

constraints are designed for the situation that each

line runs at different cycle times Therefore, these

constraints also ensure that the cycle time for a

common workstation is the minimum of the cycle

times of multiple assembly lines that it spans

Equations (5) and (6) determine whether workstation

jis utilised on line h or not If workstation j is utilised

on line h then Uhk will be 1, otherwise it will be 0

Connecting two assembly lines with any connectivity

type may not always be possible Therefore, Equation

(7) is used to disconnect two assembly lines that cannot

be connected with one or more common workstations

Equations (8), (9) and (10) are added to the model to

establish perpendicular connections For a given (a, b)

perpendicular connectivity relationship, Equation (8)

determines whether any common workstation j

be-tween lines a and b is utilised or not Equation (9)

ensures that at most one common workstation can be

utilised between lines a and b If a common

work-station between lines a and b is utilised, Equation (10)

guarantees this workstation contains the last task (k)

of the upstream line a and the task (l) of the

downstream line b to which the outputs of upstream

line a are the input Equation (11) determines whether

workstation j is utilised or not

4 Illustrative examples

The proposed model is illustrated using a simple

multiple straight assembly line facility The example

facility consists of four assembly lines with a total of

15 tasks Task completion times and cycle times of

assembly lines are given in Table 1

Suppose assembly line managers conveyed thatthere are several connectivity opportunities betweenfour assembly lines These opportunities are given inTable 2

The set of disconnection relationships can be easilyobtained using Table 2 Using these connectivityopportunities, the integrated precedence diagram isconstructed and presented in Figure 7

At the first stage, each assembly line is balancedindependently supposing no connectivity betweenassembly lines is allowed The results for independentline balances will show us the effects of commonworkstations on total number of workstations utilised

in the production facility The maximum number ofworkstations (Kmax) is selected as 6 and the problemsare solved using LINGO (Schrage 2002) on an IntelCore 2, 2.00 GHz, 1022 MB Ram computer Indepen-dent line balances are given in Table 3

Table 3 shows that a total of seven workstationsare required for the optimal solution For workstation

1, 2/5¼ 40% of total available workforce is ductive The ratios of unproductive times for otherworkstations are 40%, 20%, 50%, 30%, 40% and25% respectively with an average of 35% This can beconsidered as an important waste of workforceresources According to the assignments given in

unpro-Table 1 Task data and cycle times of assembly lines

Lines (h) Tasks (i) Task Times (ti) Cycle Times (Ch)

Table 3, the layout of the facility can be illustrated as

shown in Figure 8

The example facility is then balanced considering

the connectivity opportunities given in Table 2 Since

the cycle times of assembly lines vary, the cycle time for

a common workstation will be the minimum of the

cycle times of multiple assembly lines that it spans In

addition to the precedence relationships for each

individual precedence diagram, (6, 14) and (10, 11)

precedence relationships are added to S for reflecting

the integration of precedence diagrams The results are

given in Table 4

Table 4 shows that total number of workstations is

reduced to six when assembly lines are connected to

each other with common workstations Workstations

1, 3 and 5 are common for two different assembly lines

Workstation 1 is common for lines 1 and 3,

work-station 3 is common for lines 3 and 4, and workwork-station

5 is common for lines 2 and 4 According to the

assignments given in Table 4, the layout of the facility

can be illustrated in Figure 9

The cycle times for these common workstations are

the minimum of the cycle times that they span For

example, the cycle time for line 2 is 10 min and the

cycle time for line 4 is 20 min However, the cycle time

for workstation 5 that is common for these lines is

10 min Owing to the timing difficulties and operator/

machine interference, the multi-line (common)

work-stations that were utilised on these lines are more

difficult to operate when the cycle times of the lines are

varied (Miltenburg 1998) Consider workstation 5 in

Figure 10 as an example Task 6 must be completed

once every 10 min, while task 14 must be completed

once every 20 min Therefore, a task sequence that the

operator will follow must be generated to operate this

workstation synchronously Three Gantt charts for

task sequences of common workstations 1, 3, and 5 are

generated and given in Figure 10

The Gantt charts show when the operator must

start and finish the tasks and when the operator is idle

The task sequences in Figure 10 are generated for a

limited duration This duration is the least common

multiple of the cycle times of associated assembly lines

For instance, the task sequence of workstation 3 is

generated for 60 min because the least commonmultiple of 15 and 20 is 60 After 60 min, the operatorwill follow the same task sequence Therefore, themagnitude of workloads and idle times for commonworkstations must be calculated by considering least

Table 3 Independent line balances

common multiple of cycle times Figure 10 shows that

6 min of 15 min is idle for workstation 1, 6 min of

60 min is idle for workstation 3 and 9 min of 20 min is

idle for workstation 5 This means that 40%, 10% and

45% of available time of workstations are

unproduc-tive respecunproduc-tively The workloads of regular (single-line)

workstations 2, 4 and 6 are 8, 4 and 18 min respectively

with 20%, 20% and 10% unproductive idle time

ratios These values mean that the average rate of

unproductive times for the example multiple assembly

line facility is 24.16% That is, unproductive times for

workstations are reduced significantly compared with

independent line balances

5 ConclusionsThe main objective of the TPS is to increaseproductivity by eliminating or reducing all kinds ofwastes in a production system Shojinka is animportant TPS technique that aims to eliminate orreduce operators’ idle times If the cycle time issufficiently large, an operator can perform operations

on two or more different production lines at the sametime The problem of assigning operations to operators

is important while implementing Shojinka In thispaper, we proposed that if practical, Shojinka can beimplemented in a multiple straight assembly line

Figure 10 Task sequences for common workstations

facility The problem of assigning assembly tasks of

multiple assembly lines to minimum number of

workstations is called MSLB The MSLB problem is

built on the concept that it should be possible for a

company to obtain the advantages of Shojinka even if

the company has not adopted the U-shaped line

layout Three connectivity types to combine multiple

straight assembly lines are investigated and explained

A binary formulation for MSLB is developed based

on the connectivity types investigated The proposed

formulation assigns some tasks to common

work-stations which include tasks on two or more

neighbour assembly lines The utilisation of common

workstations minimises total idle times of assembly

lines and total number of workstations required in the

facility As with other types of ALB problems, the

MSLB problem is also NP-Hard Therefore

develop-ment of effective heuristics to solve an MSLB

problem is an important topic for future researches

In addition, the proposed binary formulation is

expected to help researchers to develop formulations

for combining multiple U-lines

References

Aase, G.R., Olson, J.R., and Schniederjans, M.J., 2004

U-shaped assembly line layouts and their impact on labor

productivity: An experimental study European Journal of

Operational Research, 156 (3), 698–711

Baybars, I., 1986 A survey of exact algorithms for the simple

line balancing problem Management Science, 32 (8),

909–932

Becker, C and Scholl, A., 2006 A survey on problems

and methods in generalised assembly line balancing

European Journal of Operational Research, 168 (3), 694–

715

Boysen, N., Fliedner, M., and Scholl, A., 2007 A

classifica-tion of assembly line balancing problems European

Journal of Operational Research, 183 (2), 674–693

Chiang, W.C., Kouvelis, P., and Urban, T.L., 2007 Linebalancing in a just-in-time production environment:balancing multiple U-lines IEEE Transactions, 39 (4),347–359

Erel, E and Sarin, S.C., 1998 A survey of the assembly linebalancing procedures Production Planning and Control, 9(5), 414–434

Erel, E., Sabuncuoglu, I., and Aksu, B.A., 2001 Balancing ofU-type assembly systems using simulated annealing.International Journal of Production Research, 39 (13),3003–3015

Ghosh, S and Gagnon, J., 1989 A comprehensive literaturereview and analysis of the design, balancing andscheduling of assembly systems International Journal ofProduction Research, 27 (4), 637–670

Go¨kc¸en, H., Agpak, K., and Benzer, R., 2006 Balancing ofparallel assembly lines International Journal of Produc-tion Economics, 103 (2), 600–609

Go¨kc¸en, H and Agpak, K., 2006 A goal programmingapproach to simple U-line balancing problem EuropeanJournal of Operational Research, 171 (2), 577–585.Miltenburg, J and Wijngaard, J., 1994 The U-line balancingproblem Management Science, 40 (10), 1378–1388.Miltenburg, J., 1998 Balancing U-lines in a multiple U-linefacility European Journal of Operational Research, 109(1), 1–23

Monden, Y., 1993 Toyota production system Norcross, GA:Engineering and Management Press

Salveson, M.E., 1955 The assembly line balancing problem.Journal of Industrial Engineering, 6 (3), 18–25

Scholl, A and Klein, R., 1999 Ulino: Optimally balancingU-shaped JIT assembly lines International Journal ofProduction Research, 37 (4), 721–736

Schrage, L., 2002 LINGO release 8.0 LINGO System Inc.Sennott, L.I., Oyen, M.P.V., and Iravani, S.M.R., 2006.Optimal dynamic assignment of a flexible worker on anopen production line with specialists European Journal

of Operational Research, 170 (2), 541–566

Sparling, D., 1998 Balancing Just-In-Time production units:the N U-line balancing problem Information Systems andOperational Research, 36 (4), 215–237

Urban, T.L., 1998 Optimal balancing of U-shaped assemblylines Management Science, 44 (5), 738–741

Job shop scheduling by pheromone approach in a dynamic environment

A simulation environment developed in ARENA1

package was used to implement the approaches and evaluatethe performance measures The performance measures investigated are: throughput time, throughput, Work InProcess, machines average utilisation and tardiness Several scenarios are considered: from static to very dynamicconditions for internal and external exceptions of the manufacturing system The simulation results highlighted thatthe performance of the proposed approach are comparable with the benchmark when the customer demand has ahigh fluctuation and the manufacturing system is less dynamic

Keywords: dynamic scheduling; ant colony intelligence; pheromone; multi-agent systems; discrete event simulation

1 Introduction

The environment within which manufacturing systems

operate is characterised by more rapid change than

ever before The unforeseen disturbances occur

fre-quently and the manufacturing system has to be able to

react to these disturbances The disturbances can be:

machine breakdowns, demand variability and delay in

processing time The above reasons drive the

manu-facturing systems to adopt dynamic scheduling

approaches

Dynamic scheduling has been defined under four

categories (Mehta and Uzsoy 1999, Vieira et al 2000a,

2003;, Aytug et al 2005, Leus and Herroelen 2005):

(a) completely reactive scheduling; in this case no

schedule in advance is created and the job

schedule is obtained in a real-time fashion

(b) predictive-reactive scheduling; a schedule is

created in advance, but a rescheduling is

considering to respond to exceptions in real

time

(c) robust predictive-reactive scheduling; a schedule

is created in advance and rescheduling activity

is activated when the effect on the performance

measures of the exceptions is significantly

(d) robust pro-active scheduling; the schedule is

computed in advance predicting the effect of

exceptions on the manufacturing system

The dynamic scheduling problem can be solved byusing the following techniques:

(a) mathematical programming approach;

(b) dispatching rules approach;

(c) heuristic approach;

(d) artificial intelligence approach

Moreover, most of the scheduling systems developed

in industrial environments are centralised and archical, but these approaches present several draw-backs The most important drawbacks are thefollowing:

hier-(a) a central computer; it constitutes a bottleneckwith a limit of capacity and a failure of it leads

to bring down the entire manufacturing system;(b) it is more difficult to extend and modify theconfiguration of the scheduling system;(c) a slow response to disturbance because theinformation has to flow to the high level andthen a reaction is planned

Therefore, a centralised and hierarchical scheduling isinefficient in a very dynamic environment where theexceptions are more frequent

For the above reasons, a decentralised controlmethod is more efficient in a dynamic environment in

*Email: paolo.renna@unibas.it

Vol 23, No 5, May 2010, 412–424

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

Ó 2010 Taylor & Francis

DOI: 10.1080/09511921003642170

particular, Multi-Agent System (MAS )methodology is

more suitable for the implementation of decentralised

system MAS can be defined as a network of problem

solvers that work together to solve problems that are

beyond their individual capabilities (O’Hare and

Jennings 1996)

MAS approaches are more suitable to develop agile

and robust distributed control, but its performance

depends more on the coordination mechanism The

coordination mechanisms proposed in literature are

the Contract net protocol (Smith 1980), market-based,

auction-based (Siwamogsatham and Saygin 2004) and

game theory These approaches have same limitations

as: communication overhead, constantly exchange of

information and therefore a minor reactivity of the

agents

Recently, many authors have developed several

approaches inspired by the behaviour of social insects

such as ants, bees, termites and wasps to propose an

alternative method for coordination in complex

systems The most promising approach is based on

Ant Colony; Ant Colony coordination refers to the

cooperative ant foraging behaviour Ant colonies can

always find shorter paths from a nest to a food source

by pheromone trail laying and following Dorigo et al

(1991) first introduced Ant Colony Optimisation

(ACO) for solving the Travelling Salesman Problem,

which is based on ant foraging In manufacturing

scheduling problem, the approach can be inspired to

pheromone-based rule of Ant Colony to coordinate the

MAS architecture

The aim of this paper is to investigate the

performance of a pheromone approach for cellular

manufacturing systems in a very dynamic

environ-ment The pheromone approaches proposed are two:

one based on moving average and the other one on

exponential moving average

Two different types of disturbances are considered

(1) Internal exceptions; these exceptions are caused

by the resources of the manufacturing system,

in particular, they are: machine breakdowns

and efficiency of the manufacturing machine

(2) External exceptions; these exceptions are

caused by external changes of the

manufactur-ing system, specifically they are: changes in

volume and mix demand products

Moreover, it is considered the effect of dispatching rule

on the performance of the manufacturing system The

performance ares compared with an approach

pro-posed in literature based on classical coordination

mechanism in MAS

The structure of the paper is as follows: in

Section 2 an overview of literature is presented The

manufacturing context is described in Section 3 InSection 4 the proposed coordination approaches based

on pheromone are explained The simulation ment and design of experiment conducted are given inSection 5, while in Section 6 the simulation results arepresented Finally, conclusions and a future researchpath are presented in Section 7

environ-2 Literature reviewRecently, a few researchers have developed approachesbased on Ant Colony inspired scheduling in shop floorcontrol Most of the applications proposed concern theshop floor routing and permutation flow-shop sequen-cing problem

Peeters et al (2001) has developed a conceptcontrol system based on coordination mechanism ofinsect colonies, in particular, a pheromone-basedcontrol scheme is introduced Basic principles of thepheromone concept, the control system architectureand a layered approach for decision-making havebeen discussed Test beds of industrial scale havebeen used to demonstrate properties and benefits

of this approach The main advantages that emergedfor the pheromone approaches are: a simple co-ordination mechanism; the automatic guidance to theoptimised solution; the capability to handle dynamicsituations

Yu and Ram (2006) proposed a multi-agentapproach designed for dynamic job shops with routingflexibility and sequence-dependent setup A bio-in-spired strategy based on division of labour in insectsocieties is presented for coordination among agents.The strategy is accomplished using a computationalmodel, which is composed of response threshold,response intention, and machine-centred reinforcementlearning The bio-inspired scheduling is compared with

an agent-based approach and a dispatching rule-basedapproach The experiments were performed usingsimulation and statistical analysis Results show thatthe proposed bio-inspired scheduling model performsbetter than the other two methods on all eight commonscheduling metrics

Xiang and Lee (2008) presented research ing dynamic scheduling by agent coordination that isinspired on Ant Colony in MAS The proposedapproach is tested in a shop floor model considersmultiple job types and parallel multi-purpose machineswith dependent setup times Moreover, the machinebreak down is included as a disturbance of themanufacturing system In this research, the AntColony is used to find an appropriate machine agentfor processing and helps the machine agent todetermine the next job to be processed in the currentqueue

concern-Scholz-Reiter et al (2008) proposed a

pheromone-based autonomous control method to a mix model of a

shop floor and tested it in different dynamic demand

situations They compared the pheromone approach

with a method based on queue length estimator The

only performance is the throughput time with a

sinusoidal arrival rate of the parts

Zhou et al (2009) proposed an algorithm based on

Ant Colony Optimisation in a shop-floor scenario with

three levels of machine utilisation, three different

processing time distributions, and three different

performance measures for intermediate scheduling

problems The performances measured are: mean

flow time, mean tardiness, total throughput on

different experimental environments compared with

those from dispatching rules including first-in-first-out,

shortest processing time, and minimum slack time The

experimental results show that ACO outperforms

other approaches when the machine utilisation or the

variation of processing times is not high The

procedure proposed is a centralised approach

Renna (2009) developed two pheromone

ap-proaches for the job shop scheduling problem One

is based on the past information of a part

(throughput time of the manufacturing cell); and

the other on the queue of the manufacturing cell

The proposed approaches are tested in dynamic

environment; the simulation results show how the

approach based on the queue of the manufacturing

cell performs better when the environment conditions

are very dynamic

Based on the above literature review, the following

limitations can be drawn:

(a) a benchmark based on MAS architecture with

complex coordination mechanism was not used

to evaluate the advantages and disadvantages

of Ant Colony coordination mechanisms;

(b) the proposed approaches in literature are tested

in manufacturing systems in which some

exceptions occur and the analysis on the

rapidity of alterations was not investigated

(c) each model proposed is tested on few

perfor-mance measures (for example only throughput

performance)

The main findings of the research proposed in this

paper can be summarised as follows:

(a) the pheromone approach is deeply investigated

by an adequate number of performance

measures;

(b) the simulations have been conducted in a

several dynamicity levels introducing both

internal and external disturbances;

(c) the pheromone approaches have been pared with a complex MAS approach in order

com-to find out in which environment conditions theproposed approach can be competitive

3 Manufacturing system contextThe manufacturing system consists of a given number

of cells; each cell is able to perform any kind ofmanufacturing operation so that the resulting manu-facturing system is a pure general-purpose one Insuch a system, the scheduling decision consists indeciding in what manufacturing cell the part willperform the next operation, therefore it is a puredispatching problem The manufacturing system hasthe following characteristics:

(1) Several part types have to be manufactured bythe manufacturing system Each part type has apredefined number of operations performed bythe manufacturing cells At each part isassigned a due date

(2) Orders for production of different parts arriverandomly with an inter-arrival that is exponen-tial distribution;

(3) Each machine performs the manufacturingoperation with an efficiency, which sets thespeed of the operation

(4) The queues are managed by the First In FirstOut policy in order to investigate only thepheromone approaches policy

(5) Each machine can breakdown randomly with

an exponential distribution

In this research, the transportation time of the materialhandling devices are included in the processing time,and the handling resources are always available.The coordination approach based on Ant Colony isinspired by the behaviour of foraging ants that leave apheromone trail on their way to the food In the realworld, ants (initially) wander randomly, and uponfinding food return to their colony while laying downpheromone trails If other ants find such a path, theyare likely not to keep traveling at random, but insteadfollow the trail, returning and reinforcing it if theyeventually find food Over time, however, the pher-omone trail starts to evaporate, thus reducing itsattractive strength

Ant Colony is a natural example of a highlydistributed system, therefore, it can be formalised byMAS The coordination mechanism based on AntColony has the following advantages

(1) The simplicity of the coordination mechanism

In fact, the ants do not communicate directly

each other, but the communication is

per-formed by the pheromone released on the

environment The ants have to know how to

put information and how to obtain information

from the environment

(2) The evaporation of the pheromone is the

methodology by which this coordination

ap-proach is capable of handling dynamic

situations

The drawbacks of pheromone approaches are the

following:

(1) The delay of the information; the ants release a

pheromone based on its activity; the other ants

put this information with a delay time

(2) The evaporation rate is a parameter to set,

therefore an opportune study on the specific

problem modelled has to be conducted

In a manufacturing system the ants are the parts that

flow through the manufacturing cells; the

manufactur-ing cells are the nodes that ants must visit When a part

leaves a manufacturing cell, it deposits a pheromone

based on the throughput time in the manufacturing

cell

Therefore, there are two main problems:

(a) how the pheromone information is formalised,

what type of information can be included;

(b) the methodology of pheromone evaporation

The activities of Ant Colony algorithm for scheduling

in manufacturing systems are the following:

(a) initialising pheromone for each manufacturing

cell;

(b) update pheromone for each manufacturing cell;

(c) pheromone evaporation

4 Pheromone approach

In this paper, two pheromone formulations have been

proposed: the first is based on moving average of the

information deposited by the parts; the second is based

on exponential moving average The two formulations

are an adaptation of the approaches proposed,

respectively in Scholz-Reiter (2008) and Renna

(2009) The main difference between the two

ap-proaches are the data stored; in moving average the

amount of data for the computation is higher than the

exponential moving average

The information deposited by the parts for the

pheromone computation is the throughput time of

a part (ant) that is processed on a manufacturing cell(node)

In particular, the value deposited by a part iscomputed by the following expression:

where,tnow is the time when the part leaves themanufacturing cell;

tin is the arrival time in the manufacturing cell ofthe part

From the above expression, the difference (tnow-tin)concerns the throughput time of the part in themanufacturing cell Therefore, if the throughput time

is low, then the pheromone deposited by the part is high.The value of pheromone at start time of thesimulation is obtained by the following expression:

where,processing time is the value of processing timeperformed by the manufacturing cell

4.1 Pheromone formulation IThe evaporation strategy of this approach is obtained

by the computation of moving average of the valuecomputed in expression 2 The pheromone is computed

by moving average over last N value deposited by theparts (expression 2) The older values deposited are notincluded, then this emulates the evaporation of thepheromone

pheromonejðtÞ ¼ 1

N

XN i¼1

Therefore, according to above description, the ities are the following:

activ-(a) initialise pheromone (expression 2);

(b) update pheromone (expression 1);

(c) pheromone evaporation (expression 3).The probability to select a manufacturing cell bythe part agent is in direct ratio to the value ofpheromonej(t); specifically, the probability to select amanufacturing cell j is

pheromonejðtÞP

jpheromonejðtÞ;with j¼ 1, , number of manufacturing cells

This formulation of pheromone is a look-back

methodology, because the pheromone value is

com-puted by past events For example, if a machine is in

breakdown status, the pheromone is reduced with a

delay time that the other parts can ‘‘sniff’’

4.2 Pheromone formulation II

The difference from the first formulation is the

evapora-tion strategy based on exponential moving average The

pheromone is computed by the following expression:

pheromonejðtÞ ¼ ð1  aÞ

pheromonejðt  1Þ þ a  value ð4Þ

The above expression is the average between the last

value memorised (pheromonej(t71)) and the value

computed by the part that leaves the manufacturing

cell (value)

The value a (it is in range [0,1]) emulates the

evaporation of the pheromone In particular, a high

value of a leads to high evaporation because the value

is more significant and vice versa for a low value of a

Therefore, according to the above description, the

activities are the following:

(a) initialise pheromone (expression 2);

(b) update pheromone (expression 1);

(c) evaporation pheromone (expression 4)

4.3 The efficiency-based approach (benchmark Renna

et al (2001))

Briefly, in this approach the MAS is composed of two

types of agent: part agent and manufacturing cell

agent The productive function of the manufacturing

cell agent consists in evaluating and providing to the

part agent the following three parameters: Expected

Part Throughput Time (ETT); Resource Failure Index

(RFI); Resource Processing Time Index (RPTI) The

ETTis computed by the following expression:

StdFlowTimekðtÞ ¼ medwtime

medwtimeþ FlowTimekðtÞ ð5Þ

medwtimeis the expected processing time of the parts

FlowTimek(t) is the expected manufacturing cell

kth throughput time computed by summing up the

processing times of the parts waiting in the

manufac-turing cell queue plus the residual service time of the

part being worked in the machine at the negotiation

time t This index is the measure of workload of the

generic manufacturing cell The index value is one if no

parts are in the queue and the resource is in idle state; it

decreases with the increase of parts in the queue TheRFIis computed by the following expression

Each parameter is normalised for each k-thresource by the following expression:

Normalised IndexkðtÞ ¼maxkindexðtÞ  indexkðtÞmaxkindexðtÞ  minkindexðtÞ

ð7Þ

Then, each index is normalised between the maximumand minimum values of the indexes among all themanufacturing cells

The above indexes are composed by the followingexpression:

Eff Index¼w  Norm EPPT þ ð1  wÞ

 ðNorm RFI þ Norm RPTÞ ð8ÞThe part agent selects the manufacturing cell that hasthe highest value of efficiency In this paper the value

of weight w¼ 0.4 that leads to better performance.The above approach needs to implement a Multi-agent infrastructure with an exchange of informationbi-directional and with a computation of this informa-tion Therefore, the approach is more complex in terms

of infrastructure and computational time

5 Simulation environment

In order to evaluate the performance of the proposedapproaches, a simulation environment is developed.The author selected the Arena1discrete event simula-tion platform by Rockwell Software Inc It was used todevelop the simulation model of the presentedapproaches The manufacturing system consists offour general-purpose manufacturing cells that arecalled to manufacture a set of four different parts Inorder to emulate a dynamic environment the proposedapproaches have been tested through a production runconsisting of several alternating stages; each stage ischaracterised by different internal (efficiency of themanufacturing cells, and machine breakdowns) andexternal attributes (mix and inter-arrival demand).The processing time of a manufacturing cell ischaracterised by an efficiency parameter (eff); in

particular, the processing time is obtained by the

following expression:

processing time¼ effj ePT ð9Þ

where, ePT is the expected Processing Time

(medw-time) for all the parts equal to 30 minutes

Table 1 reports the experimental classes tested; the

parameters concerning:

(1) Manufacturing operations; it is the number of

manufacturing operations that the parts need

The classes from 1 to 24 are characterised by

the number of operations equal for all the

parts, while from 25 to 39 classes the

manu-facturing operations are different for the parts

and the values are reported in Table 3

(2) Mix; it is the mix of the four typology parts

reported in Table 4

(3) Internal disturbance; it is the static or dynamicchange of machine breakdowns and efficiencyparameters The values are reported in Table 2.(4) Inter-arrival time dynamic; it is the dynamicchange of the inter-arrival time of the parts (seeTable 5)

(5) Congestion level; it is the level of workload ofthe manufacturing system In the case ofdynamic inter-arrival time the congestion level

is variable over the simulation length

(6) Dynamicity level; it is the rapidity changing ofdynamic conditions (see Table 6)

The evaporation parameter of the pheromone approach

I is carried out by the computation of the last fivevalues deposited The simulations conducted fordifferent values of N (see Equation (3)) highlight thatfive is the number of N that leads to better results

Table 1 Experimental classes

Exp No

Manufacturing

Internaldisturbances

Inter- arrivaltime dynamic

CongestionLevel

Dynamicitylevel

2 One for all parts Equal for all parts Static No Medium –

5 One for all parts Equal for all parts Dynamic No Low Medium

6 One for all parts Equal for all parts Dynamic No Low High

7 One for all parts Equal for all parts Dynamic No Medium Low

8 One for all parts Equal for all parts Dynamic No Medium Medium

9 One for all parts Equal for all parts Dynamic No Medium High

10 One for all parts Equal for all parts Dynamic No High Low

11 One for all parts Equal for all parts Dynamic No High Medium

12 One for all parts Equal for all parts Dynamic No High High

13 One for all parts Equal for all parts Dynamic Yes Variable Low

14 One for all parts Equal for all parts Dynamic Yes Variable Medium

15 One for all parts Equal for all parts Dynamic Yes Variable High

16 Two for all parts Equal for all parts Dynamic Yes variable Low

17 Two for all parts Equal for all parts Dynamic Yes variable Medium

18 Two for all parts Equal for all parts Dynamic Yes variable High

19 Three for all parts Equal for all parts Dynamic Yes variable Low

20 Three for all parts Equal for all parts Dynamic Yes variable Medium

21 Three for all parts Equal for all parts Dynamic Yes variable High

22 Four for all parts Equal for all parts Dynamic Yes variable Low

23 Four for all parts Equal for all parts Dynamic Yes variable Medium

24 Four for all parts Equal for all parts Dynamic Yes variable High

31 Different Dynamic changes Dynamic Low fluctuation variable Low

32 Different Dynamic changes Dynamic Low fluctuation variable Medium

33 Different Dynamic changes Dynamic Low fluctuation variable High

34 Different Dynamic changes Dynamic High fluctuation variable Low

35 Different Dynamic changes Dynamic High fluctuation Variable Medium

36 Different Dynamic changes Dynamic High fluctuation variable High

37 Different Dynamic changes Dynamic Low fluctuation With EDD Low

38 Different Dynamic changes Dynamic Low fluctuation With EDD Medium

39 Different Dynamic changes Dynamic Low fluctuation With EDD High

The evaporation parameter of the pheromone

approach II a is fixed to 0.5 in order to evaluate with

the same importance between the last value stored and

the new pheromone deposited by the part

It has been assumed that all the manufacturing cells

are subject to faults, and failures occur in accordance

with exponentially distributed time between failures,

with mean time between failure (MTBF) Repairing

times are also exponentially distributed with mean time

to repair (MTTR)

Table 2 reports the parameter of the exponential

distribution of MTBF and Efficiency (Eff) for each

manufacturing cell and each stage As the reader can

notice, the MTBF and Eff decrease the performance of

the manufacturing cell in order to emulate a

degrada-tion of the machines

The MTTR is related to the processing time by the

following expression:

Mean Time To Repair ¼ 1.5*ePT

Table 3 reports the number of visits of the fourtypology parts in order to investigate the performanceover the number of visits and for a different number ofvisits for the typology parts

Table 4 reports the mix changes over the stages ofthe simulation

Parts enter the system following an exponentialarrival stream; the inter-arrival times (exponentialparameter) are computed for the different experimentalclasses in order to keep the three levels of workload(congestion) reported in Table 1 The workload indexallows the same congestion level (high, medium andlow) to be kept when the other experimental conditionschange; it is computed as follows:

workload index¼ inter arrival  4

number of visit 30 ð10Þ

where,inter-arrival is the parameter of the exponentialarrival stream (excepted value);

number of visit is reported in Table 3;

4 is the number of manufacturing cell;

30 is the expected processing time

Then, the inter-arrival time parameter is computed

in order to obtain a value of expression 10 respectively:(a) workload index¼ 1.2 for low congestion level;(b) workload index¼ 1 for medium congestionlevel;

(c) workload index ¼ 0.8 for high congestionlevel

Finally, the dynamic change of the inter-arrival time isreported in Table 5, in which low and high fluctuation

of the inter-arrival parameter are considered To model

Table 3 Number of visit

Number of visit (table 1) Part 1 Part 2 Part 3 Part 4

Table 4 Mix changes

Stage 1 Stage 2 Stage 3 Stage 4 Stage 5

Stage2

Stage3

Stage4

Stage5

Dynamic (workloadindex)

Low fluctuation[minutes]

High fluctuation[minutes]

Table 6 Experiment classes

Dynamicitylevel

Stage lengthfactor

Simulationrun length

Stagerepeat

a highly dynamic demand, the inter-arrival time is

oscillating between two values with situation of over

and under load Table 5 reports two levels of

fluctuation of the inter-arrival time

Moreover, a specific due date has been assigned to

each part entering the system Using tjn8wto denote the

arrival time of a specific part j, its due date follows a

uniform distribution in the range given in Equation (11)

The objective is to investigate the difference among the

proposed approaches and benchmark, therefore the range

in Equation (11) has minor importance

If the part leaves the system after this due date, a

penalty is computed equal to such a delay (tardiness)

ddj¼tnowj þ UNIFORM½3; 8

where OpNumb is the number of visits of the specific

part j

The proposed approaches are tested in static and

highly dynamic situations; the dynamicity of the

manufacturing system is characterised by the stage

length factor The simulation length is fixed to 14,400

time units in order to avoid transitory influence

During the stage length the characteristics of the

manufacturing system are constant, therefore the rapid

succession of stages control the dynamicity of the

manufacturing system conditions The length of the

stage is related to the expected processing time by

the stage length factor:

stage length¼ stage length factor  ePT ð12Þ

For each experimental class (except for static classes)

the simulations are conducted for three levels of

dynamicity as reported in Table 6 The level low leads

to repeat each stage once, level medium leads to repeateach stage twice and level high leads to repeat eachstage three times

Furthermore, because of the parameters extracted

by the exponential distribution (inter-arrival time,machine breakdown), and in order to guarantee astatistical validity of the results, for each run, thenumber of executed replications guarantees, for theoutput performance measures, that the length ofconfidence intervals (95% level) of the mean amongreplications is lower than 5% of the mean itself Theperformance measures investigated are: throughputtime, throughput, Work In Process, machines averageutilisation and tardiness

6 Simulation resultsThe simulation results are reported in the followingfigures The figures report the percentage difference ofthe proposed pheromone approaches compared with-efficiency approach (base approach for determiningpercentage difference) The pheromone approaches Iand II are indicated respectively as ‘mob’ and ‘exp’ inthe following figures

Figure 1 reports the average flow time in static anddynamic conditions The proposed approaches lead tobest performance when the system is in high utilisationstate with internal exceptions (Exp No 10, 11 and 12see Table 1), and the exponential approach isalways the best approach

Concerning the throughput, the difference amongthe approaches is always under one percent Therefore,the three approaches (efficiency and the two phero-mone formulations) have the same level of throughput

Figure 1 Average flow time with one manufacturing operation

The Work In Process measure has the same trend of

the flow time performance

The average utilisation of the manufacturing

system is the same among the different approaches

Figure 2 reports the tardiness performance Better

performance is obtained when the system is in high

congestion level with internal exceptions, however the

pheromone approaches lead to a great reduction of

this performance

All the performance measures worsen when the

dynamicity increases and the congestion level

de-creases Therefore, the proposed approaches are

suitable to be used in condition of low or medium

dynamicity of internal exceptions and when thecongestion level is high The exponential movingaverage is better than the moving average

The above simulation results have been conducted

in a manufacturing system where the parts require onlyone manufacturing operation Then, the approachesproposed are tested when the manufacturing opera-tions are increasing and they are different for eachtypology part

Figure 3 shows that the flow time is the same fordifferent numbers of operations and when it isintroduced to the mix dynamicity among the typologyparts Except for the case with three manufacturing

Figure 2 Tardiness with one manufacturing operation

Figure 3 Average flow time with several manufacturing operations

operations for all the typology parts; in this case the

performance of the proposed approaches are better

than the other cases (Exp No 20 and 21) However,

the number of manufacturing operations has a low

effect on the performance The Work In Process

performance has the same trend

The throughput performance decreases in a

man-ufacturing system with three or four operations for all

the part typology, while this performance is the same

for the efficiency approach in the case of

manufactur-ing operations but different for the typology part and

with mix dynamicity (as shown in Figure 4) Then the

number of operations leads to reduce the throughput

of the manufacturing system that adopts the omone formulations

pher-Figure 5 shows that the different manufacturingoperations for the product typology and the mixdynamicity lead to the worst tardiness measure.Moreover, the performance grows worse when thedynamicity is high This confirms that the pheromoneapproaches are worst when external exceptions occur

A further exception is introduced; in particular

a fluctuation of the inter-arrival time between twovalues that leads to over and under-loading of the

Figure 4 Throughput with several manufacturing operations

Figure 5 Tardiness with several manufacturing operations

manufacturing system Two levels are considered: a

low range between the two values of inter-arrival; a

high range between the two values of inter-arrival

From Figures 6 and 7, two conclusions can be

drawn:

(1) If the oscillation of the inter-arrival time is very

wide, the performance of the pheromone

approaches are competitive with the efficiency

approach;

(2) The advantage of the efficiency approach

increases with the increase of the dynamicity

of the system

Therefore, the pheromone approaches improve

perfor-mance when the external exceptions are considerable

The Work In Process performance has the same trend

In today’s competitive environment, a high level of

delivery performance has become very important For

this reason, it is analysed how the earliest due date

(EDD) rule can improve the performance of theproposed approaches Then, the queues of the manu-facturing cells are managed by the EDD rule Figure 8shows that the pheromone approach decreases theperformance index with the EDD rule The movingaverage has a tremendous decrement of the perfor-mance, while the exponential moving average reduc-tion is limited The same trend is verified for thetardiness performance Therefore, the EDD rule leads

to reduction in all the performance measures

The same trend is verified for the tardinessperformance (see Figure 9) Therefore, the EDD ruleleads to reduction in all the performance measures andthe due date performance too (tardiness)

7 Conclusions and future developmentThe research presented a deep discussion on dynamicscheduling by pheromone approach This approach isinspired by Ant Colony Intelligence; the pheromone

Figure 6 Average flow time with inter-arrival fluctuation

Figure 7 Tardiness with inter-arrival fluctuation

Figure 8 Earliest Due Date Rule

Figure 9 Tardiness – Earliest Due Date Rule

approach is simpler than traditional approaches based

on MAS coordination (contract net, market likeness

etc.), but the performance have to be investigated A

simulation environment has been developed to test the

proposed approaches in a generic manufacturing

system The simulations have been conducted in

several conditions and in a very dynamic environment

with internal (machine breakdowns and machine

efficiency) and external (mix and inter-arrival time)

exceptions The pheromone formulation is based on

two approaches: moving average and exponential

moving average Moreover, a benchmark approach is

used to evaluate the performance measures

The results of this research can be summarised as

follows:

(1) The throughput and the average utilisation of

the manufacturing system are the performance

measures that have the same values as the

benchmark

(2) The pheromone approach is competitive when

internal and external exceptions occur and the

fluctuation of the inter-arrival demand is high

(3) The pheromone approach based on the

expo-nential moving average outperforms the moving

average; therefore the exponential moving

aver-age reacts faster to the manufacturing system

exceptions This is important, because the

exponential moving average is simpler than the

moving average for the data management

(4) A significant result is how the pheromone

approaches react to the dispatching rules In

this research the EDD rule is introduced; the

simulation results show that the moving

aver-age leads to dramatically reduced performance,

while for the exponential moving average the

reduction is minor Therefore, the dispatching

rules have to be integrated in the pheromone

computation

The reduction of the performance when the dynamicity

increases is caused by the delay information of

pheromone approach; when the dynamicity is higher

than the stage length the pheromone approach is

unable to compete with the efficiency approach

Further research paths can be the following:

(1) The influence of the manufacturing system size

on the performance measure The study of the

performance measures when a high number of

manufacturing cells characterise the

manufac-turing system An increase of the

manufactur-ing cells leads to a more complex schedulmanufactur-ing

problem; therefore a deep investigation of the

pheromone approach responsiveness to a

complexity of scheduling problem has beenperformed

(2) The extension of the research to the materialhandling system and the development of thepheromone approach for both scheduling androuting of the parts

(3) The development of hybrid approach; in thiscase, the dynamic scheduling approach can beadapted to the environment conditions There-fore, when the dynamicity is high, a complexmethod is activated, otherwise the pheromoneapproach can be performed This allows redu-cing the computational time and complexity todynamic scheduling in manufacturing systems

References

Aytug, H., Lawley, M.A., McKay, K., Mohan, S., andUzsoy, R., 2005 Execting production schedules in theface of uncertainties: A review and some future direc-tions European Journal of Operational Research, 161 (1),86–110

Dorigo, M., Maniesco, V., and Colorni, A., 1991 uted optimization by ant colonies In: Proceedings of theECAL91 European conference on artificial life, 134–142.Leus, R and Herroelen, W., 2005 The complexity ofmachine scheduling for stability with a single disruptedjob Operations Research Letters, 33 (2), 151–156.Mehta, S.V and Uzsoy, R., 1999 Predictable scheduling of asingle machine subject to breakdowns InternationalJournal of Computer Integrated Manufacturing, 12 (1),15–38

Distrib-O’Hare, G.M.P and Jennings, N.R eds 1996 Foundations

of distributed artificial Intelligence

Peeters, P., Van Brussel, H., Valckenaers, P., Wyns, J.,Bongaerts, L., Kollingbaum, M., and Heikkila, T., 2001.Pheromone based emergent shop floor control system forflexible flow shops Artificial Intelligence in Engineering,

15, 343–352

Renna, P., 2009 A performance comparison betweenefficiency and pheromone approaches in dynamic man-ufacturing scheduling In: R Chiong, ed Chapter 12 inIntelligent Systems for Automated Learning and Adapta-tion, Hershey, NewYork Information Science Reference,273–298

Renna, P., Perrone, G., Amico, M., Bruccoleri, M., andNoto La Diega, S., 2001 A performance comparisonbetween market like and efficiency based approaches inAgent Based Manufacturing environment In: 34thinternational seminar for manufacturing systems Greece.Athens, 93–98 May 2001

Scholz-Reiter, B., De Beer, C., Freitag, M., and Jagalski, T.,

2008 Bio-inspired and pheromone-based shop floorcontrol International Journal of Computer IntegratedManufacturing, 21 (2), 201–205

Siwamogsatham, V and Saygin, C., 2004 Auction-baseddistributed scheduling and control scheme for flexiblemanufacturing systems International Journal of Produc-tion Research, 42, 547–572

Smith, R.G., 1980 The Contract Net Protocol: high levelcommunication and control in a distributed problemsolver IEEE Transactions on Computers, C-29 (12),1104–1113

Vieira, G.E., Herrmann, J.W., and Lin, E., 2000 Analytical

models to predict the performance of a single machine

system under periodic and event-driven rescheduling

strategies International Journal of Production Research,

38 (8), 1899–1915

Vieira, G.E., Hermann, J.W., and Lin, E., 2003

Reschedul-ing manufacturReschedul-ing systems: a framework of strategies,

policies and methods Journal of Scheduling, 6 (1), 36–92

Xiang, W and Lee, H.P., 2008 Ant colony intelligence in

multi-agent dynamic manufacturing scheduling

Engi-neering Applications of Artificial Intelligence, 21, 73–85

Yu, X and Ram, B., 2006 Bio-inspired scheduling fordynamic job shops with flexible routing and sequencedependent setups International Journal of ProductionResearch, 44 (22), 4793–4813

Zhou, R., Nee, A.Y.C., and Lee, H.P., 2009 Performance of

an ant colony optimisation algorithm in dynamic jobshop scheduling problems International Journal ofProduction Research, 47 (11), 2903–2920

Application of concurrent engineering in manufacturing industry

Thankachan T Pullana*, M Bhasiband G Madhuca

Cochin University of Science Technology, Cochin, India;bSchool of Management Studies, Cochin University of ScienceTechnology, Cochin, India;cSchool of Engineering, Cochin University of Science Technology, Cochin, India

(Received 24 September 2009; final version received 16 January 2010)

Companies in either manufacturing or servicing have to be restructured or re-organised in order to overcomechallenges of the 21st century in which customers are not only satisfied but also delighted In this competitiveenvironment, organisations should use a flexible, adaptive and responsive paradigm Concurrent Engineering (CE) is

a management philosophy and is not restricted to manufacturing companies only It involves systematic andsimultaneous approach in developing a product or process while bringing up all the people who need to be involved

in the first place Global competitive pressure has motivated many companies to change to a more rapid form ofproduct development such as concurrent engineering (CE) By executing design in parallel, improvements occur inmany areas such as communication, quality, production processes, cash flows, and profitability Manufacturingenterprise today has become a matter of effective and efficient application of information technology andknowledge-based engineering On the one hand, this will increase the competitiveness of a firm in terms of quicklymeeting dynamic changes in the market Concurrent engineering (CE) is undertaken to improve the product designprocess with the intention of improving organisation performance

The study presented in this paper conducted an analysis of existing solutions to the problem of concurrentoptimisation of the design, and the process planning stages when a new product is developed is addressed This paperdescribes an object-oriented manufacturing process information model in the Unified Modelling Language Themodel comprises classes on the necessary manufacturing information, such as artefact, manufacturing activities,workpiece, manufacturing equipment, estimated cost and time, and manufacturing process sequences The paperadvocates for a simultaneous approach rather than the traditional sequential one

Keywords: concurrent engineering (CE); design for manufacturability (DFM); quality function deployment (QFD);computer-aided process planning (CAPP)

1 Introduction

Companies in either manufacturing or servicing have

to be restructured or re-organised in order to overcome

challenges of the 21st century in which customers are

not satisfied but also delighted Based on the study of

Cheng et al (2000), Figure 1 shows some notable

changes in manufacturing industry (Buyukozhan et al

2004) In this competitive environment, organisation

should use a flexible, adaptive and responsive

ap-proach For this, it has some useful enabling

technol-ogies and physical tools Among these, concurrent

(CE) is a systematic approach to the integrated,

concurrent design of product and their related

processes, including manufacturing and support

This approach encourages the developers to

con-sider interactively all elements of the product’s

devel-opment process from the design through to the

disposal, including customer requirements, product

quality, manufacturing costs and production time

(Lamghabbar et al 2004) Despite the wide acceptance

of the approach an implementation rate of around

50% is reported (Brookes and Blackhouse 1998) Themain barrier to implementation is the lack of tools andtechniques available to assist in implementing theapproach Quality Function Deployment (QFD) is atechnique used to implement Concurrent Engineering(Rouibah and Caskey 2005) The use of QFD for aproduct development process becomes visible with theconstruction of four Houses of Quality: ProductPlanning, Part Planning, Process Planning and Pro-duction Planning (Choi et al 2002)

Competition in the marketplace forces ing firms to continuously generate new (and moreattractive) product designs while maintaining highquality, low costs and short lead-times (Fine et al.2005) Traditionally, decisions on these issues weretaken in a serial pattern First, a product design wasselected from a set of feasible designs, driven primarily

manufactur-by marketing objectives and engineering constraints.The chosen design was then transferred to theproduction planning function that developed anappropriate manufacturing plan Such plans were

*Corresponding author Email: thankachanin@yahoo.com

Vol 23, No 5, May 2010, 425–440

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

Ó 2010 Taylor & Francis

DOI: 10.1080/09511921003643152

guided primarily by operational objectives (e.g., cost

minimisation, capacity utilisation, load balancing,

etc.) Finally, the product design and the production

plan decisions became constraints for the logistics

function that determined the supply sources This

serial pattern is known to generate solutions that suffer

from two major deficiencies (Gunasekaran et al 1994)

First, it is slow because parallel processing

opportu-nities are often missed Second, it leads to sub-optimal

solutions, because each stage can make, at best,

sequential locally optimal choices Concurrent

engi-neering (CE) is a paradigm aimed at eliminating such

flaws CE dictates that product and process decisions

are made in parallel as much as possible and that

production considerations are incorporated into the

early stages of product design The CE concept leads to

a fundamental trade-off On the one hand it reduces

the need for re-design and re-work (thus reducing

development time) and increases the chances for

smoother production (thus helping to minimise cost

and improve quality) On the other hand, CE

complicates the design problem as it requires joint

optimisation of a more complex objective with a larger

set of constraints (Wu and O’Grady 1999) Figure 2

shows processes in sequential and concurrent

develop-ment (Brookes et al 1996)

2 Related research work in manufacturing

information modelling

Most of the CE research to date has focused on

combining production considerations with product

design issues (Chang et al 1999, Gao et al 2003,

Fine et al 2005) CE applications were reported

to achieve a 30–60% reduction in time-to-market,

15–50% reduction in lifecycle costs and a 55–95%reduction in engineering change requests (Bopana et al.1997)

In design, a product model is being developed inIS0 10303 (informally known as the STandard forExchange of Product data – STEP) (IS0 1994) STEPincludes representations of geometry, topology, dimen-sion, tolerance, feature, material, product configura-tion, and so on Manufacturing information modellingefforts have been focused on manufacturing resourcecapability modelling, process plan modelling, andmanufacturing cost modelling Several manufacturingmodels have been developed, but they are not in thestandardisation stage

2.1 Manufacturing resource capability modelling

A manufacturing information model supportsthe product realisation process It only focuses on theinformation of design for manufacturability on thefactory level The two manufacturing capabilitymodels to support concurrent engineering are capable

of representing the resource capability on the station level (Noel and Brissaud 2003) A product andmanufacturing capability model for CAD/CAPPintegration focuses on information about machinetools, machining processes, operations, and cuttingtools A model of manufacturing resource informationfocuses on milling and turning machine tools, cuttingtools appropriate to the processes of milling, drilling,and so on An object-oriented manufacturing resourcemodelling for process planning includes shape cap-ability, dimension and precision capability, surfacefinish capability, and position and orientation cap-ability (Zhang et al 1999)

work-Figure 1 Development of manufacturing technology

2.2 Process plan modelling

Process plan modelling is to describe the process plan

strategy of a manufacturing process A process plan

model includes a hierarchically structured process

plan: generic plan, macro plan, detailed plan, and

micro plan A Language for Process Specification

(ALPS) has been designed as a data model to support

the description of process plans used in the discrete

manufacturing industry The design goals of ALPS

include the support for decomposition, parallel tasks,

synchronisation tasks, alternative tasks, sequences,

resource allocations, critical task sequences, and

information manipulation operatives This model is

in an entity relationship model However the model is

not object oriented STEP AP213 application protocol

(AP) within STEP supports the exchange, archiving,

and sharing of numerical control (NC) process plans

for machined parts (IS0 1994) The model supports

sequential ties It does not support parallel or

concurrent ties It also does not support manufacturingcost-time information exchange

2.3 Manufacturing cost modellingProduction costs are primarily committed at the earlydesign stage It is important to model and estimate thecosts to guide designers to make some decisions tolower product costs There are three cost-estimatingmethods used in industry: the metric-based approach(Boothroyd and Reynolds 1989, Gunasekaran et al.1994), feature-based approach (Ou-Yang and Lin

1997, Choi et al 2002) and activity-based approach(Park and Kim 1995, Rouibah and Caskey 2005).Activity-based cost (ABC) estimation is based on thecosts of all manufacturing activities ABC guideprocess plans to lower manufacturing by controllingand reducing related manufacturing activity throughidentifying non-value-adding entities The integration

Figure 2 Sequential and concurrent development of new product

of cost models with manufacturing resource capability

models and process plan models has to be further

developed to enable software interoperatability

2.4 Summary of manufacturing information

The above-mentioned models have not been fully

integrated with each other or with an information

model Some specific issue addressed are as follows:

(1) Most published process plan models found

detailed process planning, not the preliminary

process planning in the early product

develop-ment stage, and need to be extended to the

following manufacturing information

hier-archical structure of manufacturing entities,

workpiece information, processing and

manu-facturing cost

(2) Most manufacturing resource models

incorpo-rate many functional and geometric

character-istics of resources, but not the required

behavioural characteristics and capability

dur-ing processdur-ing

(3) The type of methods used for manufacturing

cost and time estimation should be integrated

into the manufacturing process model

To advance the state of the art, an integration

framework has been developed, as well as an integrated

manufacturing object model

3 Concurrent engineering

Product and process technology is rapidly evolving

Competition is becoming more and more globally

based (Wick and Bakerjian et al 1993) Customers are

placing an increased emphasis on quality and

relia-bility, but at the same time looking for good value

Time to market, or speed to market is becoming aparadigm of world class manufacturing To respond tothis increasing dynamic and challenging environment,manufacturers are implementing concurrent engineer-ing concepts to reduce design cycle time and productvalue While design for manufacturability (DFM) is acore part of concurrent engineering, its concepts arebased on an expanded focus of the entire productlifecycle from concept development through use anddisposal Concurrent engineering is based on theintegrated design of products and manufacturing andsupport process It is not a matter of assessingmanufacturability of the product after it has beendesigned and making appropriate changes to theproduct design to enhance its producibility Thisapproach extends the design cycle time, increasesproduct development cost, and may not result in themost optimum way to produce the product Instead,manufacturability must be considered from the verystart of product development and designed into theproduct The design of the product and the processmust be integrated to assure a more optimumapproach to the manufacturing of the product.Additional considerations also need to be integratedwith the design of the product A framework for thediscussion of concurrent engineering is presented inFigure 3

3.1 Strategic benefits of concurrent engineeringConcurrent Engineering (CE) is a management philo-sophy dedicated to the improvement of customersatisfaction through improved quality, reduced costsand faster product development Concurrent Engineer-ing leads to:

(a) Improved customer satisfaction(b) Improved quality

Figure 3 A framework of concurrent engineering

(c) Reduced cost

(d) Reduced new product development time

(e) Reduced time to market

(f) Reconciliation of conflicting requirements in

product development

Table 1 shows the effect of concurrent engineering on

percentage (%) product development Cycle-Time

reduction by industry, based on 25 researches studied

and reported by Paul D Collins and Alan Leong

(Chen and Hsiao 1997, Beninati 2006)

3.2 Understand the advantages: industry experience

CE has led to dramatic benefits for a large number of

companies from various industries Some of the

findings are presented here as a pointer towards the

potential benefits of this best practice (Table 2) (Harris

and Nagalingham 2000)

By executing design in parallel, improvementsoccur in many areas such as communication, quality,production processes, cash flows, and profitability Thereduction of time to market, which has strategicimportance, allows companies to increase their marketshare and reduce design changes and iterations.Product designs are more easily manufacturable,serviceable and are of higher quality Once the designsare released to manufacturing, production progressesquickly to full volume because the process is welldefined, documented and controlled

World-class companies have achieved remarkableperformance using concurrent engineering Boeing’sBallistic System Division achieved the followingimprovements:

(a) 16% to 46% in cost reduction inmanufacturing;

(b) Engineering changes reduced from 15–20 to 1–

2 drafts per drawing;

(c) Materials shortage reduced from 12% to 1%;(d) Inspection costs cut by a factor of 3

NCR, Ohio, USA used CE to develop a new cashregister and achieved (Gao et al 2003) the followingbenefits:

(a) reduction in parts and assembly line;

(b) 65% fewer suppliers;

(c) 100% fewer screws or fasteners;

(d) 100% fewer assembly tools;

(e) 44% improvement in manufacturing costs;(f) a trouble-free product introduction

Other examples are: Rolls-Royce reduced the lead-time

to develop a new aircraft engine by 30%; McDonnellDouglas reduced production costs by 40%; ITT, USAreduced their design cycle-time by 33% for itselectronics counter measuring systems Intel-Pentiumdevelopment team reconciled conflicting requirements

in semiconductor development (Kim et al 2001).Many other cases which corroborate the benefits ofadopting concurrent engineering have also beenreported

The greatest impact and benefits of concurrentengineering are realised at the design stage of productdevelopment The design decisions made in the earlyphases of product design and development will havesignificant impact upon future manufacturing andlogistical activities (Gao et al 2003) The followingexamples signify such an importance:

A study at Rolls Royce revealed that designdetermined 80% of the final production cost of 2000components ; according to General Motors executives,70% of the cost of manufacturing truck transmissions

Table 1 Effect of Concurrent Engineering on Percent (%)

Product Development Cycle-time Reduction by Industry1

.Cycle-time reduction averages about 50%.

.Robust effects among firms in many industry groups.

1

Based on 25 research studies; reported in Paul D Collins and Alan

Leong, Performance Effects of Concurrent Engineering, University of

Washington, 1998.

Table 2 Benefits obtained from concurrent engineering

Decreased lead time

Improved quality

Engineering changes 65–90% fewer

Scrap and rework up to 75% less

Overall quality 200–600% higher

Reduced cost

Return on assets 20–120% higher

Manufacturing costs up to 40% lower