04421913.pdf

LR. th.ff 2007 International Conference on Management Science & Engineering (14th)/ August 20-22, 2007 Harbin, P.R.China

Optimization Model Analysison Supply Chain Resources Integration Decision in 4PL

YAO Jian-ming, LIU Li-wenSchool of Economics and Management, Tsinghua University, P.R.China, 100084

Abstract: Integrating the supply chain resources isan important way to promote the utilization efficiency ofthe enterprises' various resources and the enterprises'operational effect and profits. But considering thecomplexity of the resources, the integrating process isdifficult. At present, the successful operation of thefourth party logistics (4PL) in practice has graduallydemonstrated that it is an effective mode to integrate thesupply chain's complicated resources relationally,efficiently and flexibly. To thoroughly analyze theintegration problem of the supply chain in 4PL modefrom the quantity view, the integration decision methodis put forward based on the dominant factors judgmentand analysis. Then a mathematics model of integrationdecision optimization is advanced and the reasonabilityand feasibility of the integrating decision method arebriefly described in this paper. Finally, the relatedconclusions of this research are given.

Keywords: 4PL, supply chain resources integration,dominant factors, decision, optimization model

1 Introduction

The supply chain resources integration process isgenerally enslaved to a paradox for a long time. Thisparadox is among "how to handle a reasonable level ofcustomer service satisfaction, the resource integrationcosts and the operating profits of the integrated supplychain system". Therefore, we must probe the way toeffectively integrate, operate and monitor the complexresource of supply chain and to maximize the current andlong-term benefits of the supply chain members. Theoperation essence and the core superiority of 4PL (thefourth party logistics) lie in its ability to integrate thesupply chain resources. Through integrating the mosthigh-quality supply chain resources (cooperators),flexibility of the supply chain will be get to a full extend.Therefore, customers will receive personalized,diversified integration solutions, and fast, high-qualityand low-cost logistics services will be providedP']. It isobvious that integration resource in 4PL mode can create

Supported by the National Natural Science Foundation ofChina (70532004) and the foundation of Education Ministry ofthe People's Republic of China (No. 06JC630017).

favorable conditions to resolve this paradox. It will begradually proved that resource integration mode of the4PL has become the dominant direction of the logisticsindustry development either for theory or in practice[24].To implement 4PL mode successfully, numerousenterprises (including IBM, Lucent, UPS) have referredto some well-known management consulting firms, suchas ACCENTURE and AXA.

In current situation, 4PL have great practicaldevelopment and good implementation environment[5],either from view of 3PL (the Third Party Logistic)development prospects, supply and demand balance, orthe system development, According to statistics, it isaccounted that in 2002 the total cost of the globallogistics expenditures has reached four trillion dollars,among which expenditure of logistics market hasachieved 281 billion dollars, accounting for 700 of totalglobal logistics. From 1996 to 2002, global logisticsmarket has grown at an average annual rate of 13%[6].

However, it should be of great concern is that therational and efficient supply chain resource integrationby 4PL mode is a complex systemic problem. Whileencouraged by successful integration operation of 4PL,we should explore 4PL integration project and strategiesat theoretical level.

Initially concept of 4PL was put forward inACCENTURE reports in 1998. Thus many subsequentresearches on 4PL were extensive discussion of thisreport. There are mainly two categories involved 4PL.One is in newspapers and magazines, mainly about thepractical application of 4PL[78]. There existed someintroductory and scholarly articles. Most of these articlesfocus on the qualitative introduction of the new conceptof 4PL[23]. These literatures are rather abundant,involving the significance and advantages of 4PLcompared with the 3PL19 '01, definition of 4PL concept,feasibility and inevitability of 4PL development[6"1112],characteristics and the operational pattern of 4PLI''13'14].

Another part studies the operation problems underthe guidance of the 4PL by the introduction ofquantitative methods. Such as the researches of theoperational problems of the supply chain in 4PL mode[14]and the technical support of the 4PL used in theoperation of the supply chain network['15 and so on. Theresults of such studies are less.

- 614 -

The existed 4PL researches have critical referencevalue to set up basic operation mode structure and guideresearch direction in this field. However, in presentliteratures, there have not been systematic researchtheories or achievements on the supply chain resourcesintegration by 4PL. In integration decision process,analysis and researches on relationship among dominantfactors, and connection between integration input andoutput are especially absent. If researches on 4PL havealways rested on qualitative concept analysis andoperational pattern discussion, expansion of the 4PL inpractical operation will be limited. Meanwhile, withoutreasonable quantitative modes of 4PL, superiority of thisoperational mode can not be thoroughly reflected. Indeedthese study tasks are very necessary and urgent.Enterprises involved in integration should not onlycomprehend necessity of their input costs, but alsounderstand relationship between costs input andimproved operational level output from quantity angle.Through these researches, devotion should be put intothe factors which will bring maximum benefits toimprove the level of its own operation system and thewhole supply chain. Therefore, quantitative mode forintegration decision at theory level should be explored.From perspective of value increment theory and systemicintegration principles, dominant factors of thisintegration decision process are analyzed at operationallevel firstly. Then to direct the practice, an optimizedmathematical model for this complex resourcesintegration is put forward and discussed in this paper.

2 Analyses of dominant factors

2.1 Importance of digging out dominant factorsAccording to system value theory, the key issue of

4PL integration is how to systematically integrate supplychain resources (including each provider of system itsown as well as individual elements of advisory membersin supply chain system) that are correspondence to acertain kind of product or service. On the premise thateach sub-systems individual will achieve its satisfiedvalue, the whole value chain should be optimized andvalue increment of operation activities be realized. Toimplement this process, it is necessary for 4PL toevaluate and select high quality resources of the supplychain. So we should determine factors used to judgeindividual integration degree above all.

Generally speaking, when 4PL integrating entitiessuch as production enterprises, commercial enterprises aswell as the 3PL individuals, resources integrationemphasis particularly is attached to improvement ofoperational level of substance flow and capital flow inproduction and service activities. When 4PL integrateentities such as management consulting individuals andinformation technology support individuals, stress shouldbe especially placed on the improvement of theoperational level of information flow. Meanwhile weshould notice that these three basic elements forproduction and service activities constitute a synthetic

operation system that is related to each other. The resultof resources integration will bring the rational flow ofobjects, capital and information and smooth linkage ofthese factors between each individual to supply chainsystem, so as to realize value increment of the system.

From above analysis it is conclude that achievementof system value increment should be directed by strategyof 4PL supply chain resource integration. On one hand,operation level of system individuals' hardenvironmental factors such as the substance flow, capitalflow and information flow should be improved. On theother hand, changes pertinent to soft environmentalfactors such as the management style, the organizationalstructure and the operational procedures should becarried out. Thus, flow of substance, capital andinformation within production and service area in systemwill joint to each other smoothly. Also operation costs ofthe major factors will decrease and operation efficiencywill improve. Meanwhile, operation level of softenvironmental factors, including management style,organizational structure and business processes of wholesupply chain members will be enhanced.

Either for principle of system nature or increment ofvalue chain, supply chain resources integration of 4PLshould be improved from the hard and softenvironmental factors that determine its individualoperation level. On the other hand, there might bedifference between each collaboration individual statuswith particular production or service activities. Hencecost put on soft and hard environment factors will bedifferent, which lead to diverse supply chain resourceintegration cost. When 4PL integration decision is made,it is the first thing we should consider. Consequently sixdominant factors that will influence the decision processof resource integration are chosen as main object foranalysis. But, just an exploratory study on 4PL mode,these dominant factors selected here are not absolute.

Besides in practical operation process, it is criticallyimportant to consider factors (such as individualreputation, value of brand integration, individual status inthe market, the individual ability to resist risk, incomelevel from the chain reflections and so on) based onlong-term benefits objectives, and satisfy basicconstraints of collaboration individuals.

2.2 Important relationship analysisDuring process for supply chain integration decision

and constitution of decision optimization model and itscorresponding algorithm, several relation issuesfollowing should be settled reasonably.

(I)To improve operational level of hardenvironmental factors and soft environmental factors aresupplement for each other. While 4PL improve systemindividual management manner, organization structureand operation procedures, with direct intention toenhance operation level of individual substance flow,capital flow and information flow. Only when systemvalue increment attain the required operation level, flowof basic production and services factors such as product

- 615 -

The cost put into the soft Management styleenvironmental factors 2

Input:Inputted cost in ,4PL integration ZE Q

The cost put into the hard c7 Logistics %environmental factors

Fig.1 The complex relationships o

/services, capital and information will become rationaland efficient. It will reduce operation cost and achievethe essential objective of 4PL. On the contrary, toimprove the operation level of various flows throughhard environmental factors (such as equipmentrenovation, new technology introduction, qualified staffimport and introduction of efficient production &services operations mode) will inevitably bringpromotion of advanced management methods andorganization models, and improvements or adjustmentsof business process. Therefore, there are the dual benefitswhen improving cost is devoted on hard environmentalfactors and soft environmental factors. In Fig. 1 the type(X) (solid line double arrow) shows the influences.

(2)Internal factors inside hard and softenvironmental factors will take their influencerespectively. In addition to above-mentionedrelationships, improvements of each insideenvironmental factors hard or soft environmental factorswill challenge operation level of any other internalfactors' and impel to promote their operation efficiency.Moreover, it will provide essential coordination andsupport to improve operation level of the other factors.As far as hard environmental factors concerned,information flow improvement by 4PL will be providenecessary technical support to improve smooth andefficient operation for substance and capital flow. On theother hand, excellent information flows also requireefficient operation of logistics and financial resourcesurgently, so as to bring harmony operation of the wholesystem. Similarly, factors within soft environmentalfactors have the same complement relations betweeneach other. Relationships between the same internalenvironmental factors are shown in Fig. 1 the type ((double dotted arrow).

(3)Improvement origin of each factor should bejudge in reason. Through above complex relationanalysis, it is approved that improvement cost put intovarious factors will have multiple effects. To accuratelyobtain cost during system integration procedure andoptimize integration cost based on improved operationlevel of supply chain system, it needs to reasonablyquantify effectiveness level resulted form cost put intothese factors. That is to say, when cost put on a factor, itshould be clearly demonstrated and quantified thatcontribution level of improvement of the factor itself (see

f do

pertio proceduOperating level of softOperation procedure environmental factorsstructure - Output:

'_________L____q__ Synthetic- L-/

~~~~~operating7lowa level of system

C Informatinflow Operating level of hardenvironmental factors

minant factors in 4PL integration

the relation () in Fig. 1), operation level of any otherinternal factors within the same environment (see therelation (@) in Fig. 1) and operation level to improvefactors of the other environment (see the relation (0) inFig. 1).

2.3 Complex input-output relationsIn addition to the mutual influence relations among

the factors, the relationship between the input and outputof the 4PL integration process is also complicated. Theinput of the process is the integration cost and the outputof it is the overall supply chain operation level.Integration cost will be imported into improving theoperational level of the hard and soft environmentalfactors by 4PL integration strategy. The integrationoutput of the system will not only be mainly reflected inthe operating state that the system's substance, capitaland information can flow and converge smoothly andefficiently along the supply chain, but also embodied inthe operating level of the system's management style,organizational structure and operational processes.

Here, we must pay attention to the problem of thequantified allocation. One is that the integration costimported into the system will observe how the ratio to beassigned to system's hard and soft environmental factorsto improve the environmental operation level and will bealong with how the ratio to be assigned to the differentfactors in the same environment. Another is what theproportional relationship among the had and softenvironmental factors' operation level contribution to thesystem's output synthesized operation level after thesystem's resource had been integrated by 4PL. Obviously,this is a complicated decision-making process, as well asthe core of 4PL operation. Therefore, in essence, thesystem integration process can be seen as the equilibriumprocess of the complex cooperation and competitionrelations between the overall system and the individualsand among the individuals themselves. So, the realizationof the system's integration must depend on thecooperation with the individuals' concerted effort, it isalso we should consider when we build an optimizationmodel of the 4PL integration.

2.4 The dynamic character of integrationThe process of the supply chain integration by 4PL

is not only a simple optimization selection problem of

- 616 -

the cooperators, but also a synthetic process coveringboth selecting the cooperator members, building the newspecific cooperator members, planning the operationscenarios, providing the information and businesssupport and managing and controlling the production andservice activities. Here, the dynamic nature of theresources integration refers to that the resources of thesupply chain are changing dynamically when the 4PLintegration is in progress. That is to say, the integrationprocess is not just aimed at existing resources but willalso create the new resources in accordance with thespecific requirements of the production and servicesactivities in order to achieve the systematic and efficientdistribution of tasks. For example, when 4PL integrationis in progress, the six-month study tracking is firstlyneeded. However, in this particular period of time thereare not suitable members to choice. Therefore we mustset up a new operation organization (system resources) tocomplete this investigation task. Clearly, it is necessaryto spend the integration cost (the cost of set up a newoperation organization) of it.

We can see that whether it is needed to set up thenew resources, what type the new resources will belongto, what the number of the new resources and how muchintegration cost we will devote to these new resourcesare to be considered in the integration decision-makingprocess. This question should also be reflected in ourintegration decision optimization model and algorithm.

3 Building of model

3.1 Variables and parameters explanationWe will build an optimization model of the resource

integration decision by 4PL based on the former analysis.In the model, we let K denote the categories of the

demanded resources for 4PL's integration aimed at someproduction or service activates. In supply chain system, itis the categories of the system's individuals which willinclude the existing individuals and the new ones; let kdenote the index of the individual categories(k-il,2,. . .,K); let Mkdenote the number of the individualswhich can be for integration in the k category; let ikdenote the index of the individual (ik =1,2,..., Mk).

We let F denote the expected synthetic operatinglevel of the supply chain system after integrated by 4PL;let FLog, Fcap and F71f denote respectively theexpected operating levels of the logistic, capital andinformation in the supply chain system after integratedby 4PL; let FMan, Forg and Fpro denote respectivelythe expected operating levels of the management style,the organizational structure and the operationalprocedure in the supply chain system after integrated by4PL; let FLogik , FCap.ik and FI denote respectivelythe operating levels of the logistic, capital andinformation of the individual ik in the supply chainsystem before integrated by 4PL; let FLo9.ik* F*ap.ik and

F* denote respectively the operating levels of thelogistic, capital and information of the individual ikin thesupply chain system after integrated by 4PL; let FMan ikIForgi and Fproi denote respectively the operatinglevels of the management style, the organizationalstructure and the operational procedure of the individualik in the supply chain system before integrated by 4PL;let Fmi, F * and F denote respectively theGi'kOg Piku'operating levels of the management style, theorganizational structure and the operational procedure ofthe individual ik in the supply chain system afterintegrated by 4PL; let AFMan.ik , AForg.ik AFPro.ik andAFLog.ik , AFcapF k /A-FInf1k denote respectively theincremental improvements of the every dominant factorsof the individual ik in the supply chain system afterintegrated by 4PL. There is also the relation:AFmanik =: Fman.k -Fman.ik and the other factors will havethe similar relationships and not to be repeated here.

We let CSoft i denote the total cost inputted into theindividual ik to improve its soft environmental factoroperation level; let ACMan ik , ACOrg ik andACprok denote respectively the cost inputted into theindividual ik to improve its management style,organizational structure and operational procedure in theintegration process; let CHard i denote the total costinputted into the individual ik to improve its hardenvironmental factor operation level; let ACLog ikIACCapCi and AC Inf i denote respectively the costinputted into the individual ik to improve its logistic,capital flow and information flow in the integrationprocess.

We let Qk denote the operating quality provided bythe individual ik; let Q denote the standard qualitydemand of the production or service category k; letAk denote the ability of the production or serviceprovided by the individual ik; let A* denote the abilitydemand of the production or service category k; letTstai denote the starting times of the production orservice activities provided by the individual ik; letTEnd i denote the ending times of the production orservice activities provided by the individual ik; let

denote the starting times demand of the productionor service category k; let Jfl denote the ending timesdemand of the production or service category k; let6sta k denote the tolerant time value of the starting timesabout the production or service category k; let 6End k

- 617 -

Tab.1 Affect relation coefficients of the factors in the integration processHard environmental factors Soft environmental factors

Yii Y21 Y31 Management style H12 H13Y12 Y22 Y32 H21 Organizational structure H23Y122 3 Y33 H31 H32 Operational procedure

Logistic G12 G13 RI, R12 R13G21 Capital flow G23 R21 R22 R23G31 G32 Information flow R3I R32 R33S11 S22 S33 V1 I V22 V33

denote the tolerant time value of the ending times aboutthe production or service category k; let 6k denote thetotal tolerant time value of the production or servicecategory k. There is the relation: 6k = 6Sta.k + 6End.k * Let,8 denote the coefficient to unify the units of the cost andthe tolerant time value. We assume if individual ik isselected to join in the 4PL integration then (3 - 1,otherwise ( =0.

Set the affect relation coefficients of theimprovement brought to the factor itself, to the otherfactors in the same environment and to the factors in thedifferent environment by the integration cost inputtedinto the dominant factors are shown in Tab. 1. Therelations in Tab. 1 follow the clockwise principle. For

example, the affect relation coefficient of the cost putinto the management style factor will bring theimprovement to the organizational structure factor isdenoted by H12; the affect relation coefficient of the costput into the organizational structure factor will bring theimprovement to the operation level of the capital flowfactor is denoted by R22. S and V denote the relationsbetween the cost inputted into the factor and theimprovement of the operation level of the factor itselfbrought by this cost.

3.2 Building of modelThe complete formulation of the model is as follows

and the objective function is

minZ=inC )6 mi OT,al =K MkminZ=minCTotal + m6- m[(CSoftlik + CHardl)k + /3k ]5kK M

E E [((ACMan.k +ACOrgik + ACProk )+ (ACLogik +ACCapik +ACInfA)) /(OSta.k + 0End.k )]ik= ik =1

Subject to:AFLog.ik (RllIACManik +R12ACOrgik +RI3ACPro.ik ) + IACLog.ik +(GI2ACCapik +G13ACInfik )AFCapi =(R21IACManik +R22ACOrgik +R23ACPro.ik ) + S22ACCap.ik +(G21IACLog.ik +G23ACInf.ik )AFjInf1 =(R31IACManik +R32ACOrgik +R33ACProik ) + S33ACInf.k +(G31IACLog.ik +G32ACCapik )

FLog.ik 2 FLogF< >FCapit CapJnf.i nf

AFManik = (Y11ACLogik +Y2 ACCapik +31ACInf )1+VIACManik +(HI 2ACorg ik +H13ACProi )AForgik =(12ACLogi +Y22AC Capi+k 32ACInf.k ) V22ACOrgk +(H21AC Manik +H23ACProik )AProik =(Y13ACLog.i +Y23ACCapi +f33ACInf1 ) +V33ACPro.ik +(H3ACManik H32COrgik )Fman it > FMan

FOrg.ik > FOrgFP*ro it 2 FproTsta 'TSat.ik |- 0Sat.k

TTEInd.k TEnd.ik Qk

E11k k Akik =1

- 618 -

(1)

(2)(3)(4)(5)(6)(7)(8)(9)

(10)(1 1)(12)(13)(14)(15)(16)

(17)

In the model, Eq. (1) is a multi-objective function,the operating main line lies in optimizing the supplychain system's integration cost and the tolerant value ofthe time limit of the production or service. Eq. (2)-(4)are the constraints of the relations between the operationlevel improvement of the hard environmental factors andthe integration cost put into every dominant factors; Eq.(5)-(7) are the constraints of the operation level of thesystem's hard environmental factors; Eq. (8) (10) are theconstraints of the relations between the operation levelimprovement of the soft environmental factors and theintegration cost put into every dominant factors; Eq.(11) (13) are the constraints of the operation level of thesystem's soft environmental factors; Eq. (14) and (15)are the constraints of the production or service timeprovided by the system individuals; Eq. (16) is theconstraint of the production or service quality providedby the system individuals; Eq. (17) is the constraint of theability provided by the system individuals.

We use an improved ant algorithm[16 18] to solve theabove model. We also validate the reasonability andfeasibility of the supply chain resources integrationdecision method in 4PL mode discussed in this paperthrough a case study and simulation.

4 Conclusion

This paper established a resource integrationdecision optimization mathematical model based on theanalysis of the dominant factors in the resourcesintegration process of the supply chain in 4PL. Themodel not only reflect the complicated basiccharacteristics of the supply chain resource integratingprocess but also merge into the solution methods ofseveral important relations in the integration process.This is also the transition bridge from the integrationdecision process to the practice. The dominant factors inreality which play a decisive role in the integrationdecision are numerous and complex and it is needed togive the deep, detailed and refined analysis. For theintegration mode of 4PL is a leading edge operationalmode in the supply chain management field. In the futurestudy (such as the papers[19'20]), we should focus on thedefinition and mining of the dominant factors.Particularly in the different production or serviceindustries, the dominant factors were most important tograsp. Meanwhile, it need more depth analysis andreasonable definition about the complex relationshipsbetween the various factors in order to guide practicewith better integration strategy and exploit the 4PL'sadvantages in the supply chain resource integration to thefull.

References

outsourcing?[J]. Logistics Management & DistributionReport, 1999, vol.38, no.435.[2]G. Marino. The ABCs of 4PLs[J]. IndustrialManagement, 2002, vol. 44, no.1:23.[3]S. Roberts. Fourth party politics[J]. FrontlineSolutions Europe, 2003, vol. 12, no. 5:14-16.[4]H. Susanne, A. Monica. Strategic development ofthird party logistics providers[J]. Industrial MarketingManagement, 2003, vol. 32, no.1:139-149.[5]Y, Li. Research on 4PL development in China[J].Business Research, 2004, no.3:169-172. (in Chinese)[6]S. Tierney. Now there are real 4PL possibilities[J].Supply Chain Europe, 2004, vol. 13, no.l:16-18.[7]H. L. Richardson. 3PL Today: A Story of ChangingRelationships[J]. Transportation & Distribution, 2002,vol. 43, no.9:38-40.[8]L. Sullivan. Distributors tangle with 4PLs for marketshare[J]. EBN, 2003, vol. 1344, no.1:1-2.[9]L. L. Sowinski. The Top 3PLs[J]. World Trade, 2000,vol.13, no.7:60-6 1.[10]J. Love. 3PL/4PL--Where Next?[J]. Logistics &Transport Focus, 2004, vol.6, no.3:18-21.[11]S. Hertz, M. Alfredsson. Strategic development ofthird party logistics providers[J]. Industrial MarketingManagement, 2003, vol.32, no.2:139-149.[12]L. Wells. 4PL goes for expansion[J]. MotorTransport, 2005, vol.1, no.1:7.[13]C. Serant. 4PLs stretching boundaries of outsourcing[J]. EBN, 2002, vol.1318, no.1:3-4.[14]J. Bumstead, K. Cannons. From 4PL to managedsupply-chain operations[J]. Logistics & Transport Focus,2002, vol.4, no.4:18-24.[15]C. L. Hoong, G. G. Yam. An intelligent brokeringsystem to support multi-agent web based 4th-partylogistics, in Proc. Conf. tools with artificial intelligence[J]. TAI'02, 154 -164.[16]E. Bonabeau, M. Dorigo, AZ. G. Theraul. Inspirationfor optimization from social insect behavior[J]. Nature,2000, vol.406, no.1:39-42.[17]N. Varela. Ant colony optimization forvirtual-wavelength-path routing and wavelengthallocation, in Proc. Conf. evolutionary computation,EC'99, Washington DC, USA, 324-337.[18]J. M. Yao, X. M. Zhang, L. W. Liu. Analysis ondynamic scheduling optimization of pull supply chainbased on improved ant algorithm[J]. Chinese Journal ofManagement Science, 2006, vol.14, no.3:20-26. (inChinese)[19]J. M. Yao, L. W. Liu. Decision and optimizationanalysis on supply chain resources integration in 4PL[M].System Engineering, in press. (in Chinese)[20]J. M. Yao, L. W. Liu. Analysis on supply chainresources integration decision in 4PL[J]. Science ofScience and Management of S. & T., 2007, vol.28,no.3:19-24. (in Chinese).

[1]T. Foster. 4PLs: The next generation for supply chain

- 619 -