Elsevier

Information Sciences

Volume 235, 20 June 2013, Pages 242-258
Information Sciences

Supply chain outsourcing risk using an integrated stochastic-fuzzy optimization approach

https://doi.org/10.1016/j.ins.2013.02.002Get rights and content

Abstract

A stochastic fuzzy multi-objective programming model is developed for supply chain outsourcing risk management in presence of both random uncertainty and fuzzy uncertainty. Utility theory is proposed to treat stochastic data and fuzzy set theory is used to handle fuzzy data. An algorithm is designed to solve the proposed integrated model. The new model is solved using the proposed algorithm for a three stage supply chain example. Computation suggests an analysis of risk averse and procurement behavior, which indicates that a more risk-averse customer prefers to order less under uncertainty and risk. Trade-off game analysis yields supported points on the trade-off curve, which can help decision makers to identify proper weighting scheme where Pareto optimum is achieved to select preferred suppliers.

Introduction

Supply chain (SC) risk management has attracted considerable and increasing attention in both industry and academia, as their activities and requirements become increasingly complex [25], [68]. This is especially true when the companies are expanding the geographical scope of their sourcing activities into areas where they have little experience as low cost country sourcing, e.g., China [81].

Outsourcing is a recent trend, usually adopted to gain lower production costs, but also can be used to reduce core organizational risk. In a global market, supply management offshore-sourcing strategies can include manufacturers at low cost locations such as China, India, or Vietnam, assemblers at high-tech operations in Taiwan and Korea, and distributors where customers reside all over the globe. They can also include e-business operations such as Amazon.com. The selection of suppliers in a global market is often considered as a problem involving complex systems. This has been severe under the supply chain management framework, because the factors such as default risk from other SC members and the effects to SC partners need to be considered. On the other hand, external stakeholders such as rating agencies affect the selection of appropriate suppliers by assigning independent, objective and non-binding opinions (not recommendations) on the financial strength of outsourcing candidates in the form of a globally consistent rating scale. The credit rating process historically includes business and financial risks of an organization in addition to indicators of macroeconomic conditions [3]. External credit ratings are thus particularly important in financial services because a higher credit rating establishes and maintains market confidence. All these indicate that factors causing SC risks, their relation and the possible effects of these risks can be very complex, making practical supply risk management in industry a difficult undertaking [36].

Since supplier selection activities under supply chain management framework are complex, a supplier selection approach must be able to take this complexity into account. Many models are available to support supplier selection and outsourcing. Ref. [42] modeled supplier risk attitude with respect to risk aversion. Some studies [2], [10], [37] have recognized this imprecision through methods accommodating fuzzy data. But these methods fail to consider uncertainty and risk factors in an integrated model. Probability distributions from historical data are widely recognized by researchers [13], [15], [24], [28], [29], [41], [65], [69] to model SC uncertainty (e.g., uncertain demand) in a decision model. However, because one single criteria such as the minimization of expected cost or maximization of expected profit is used, these decision models may result in sub-optimal solutions. A practical decision of selecting SC partners and sourcing arrangements usually exhibits as a multi-objective decision making problem [49], where multi-objective programming models have been presented [76], [77]. But existing multi-objective programming seldom simultaneously considers multiple objective and uncertainty and risk.

In this paper, a stochastic fuzzy multi-objective programming model (SFMOP) vendor selection model is developed for supply chains outsourcing risk management. We recognize that data regarding the expected performance of suppliers in a global market are necessarily imprecise. Moreover, selecting an ideal supplier is much riskier than its domestic counterpart due to a number of exogenous risk factors influencing offshore sourcing. Therefore, both quantitative and qualitative supplier selection risk factors are examined. Quantitative risk factors include cost, quality and logistics, each expressed with stochastic data with some probability distribution. Qualitative risk factors include economic environmental factors and vendor ratings, which are of a fuzzy nature and can be quantified by a degree of belief (e.g. membership function).

We model a SC consisting of three levels and use an example with simulated data extracted from our previous study. We conduct various analyses, to include sensitivity analysis on certain confidence of level (α-cut-level), simulation on weight, trade-off game analysis and two-way comparison between the proposed model and the model with three-objective case (see Section 3 for both three-objective and five-objective cases).

The rest of the paper is organized as follows. Section 2 presents literature review. Section 3 presents stochastic fuzzy multi-objective programming models. Section 4 discusses our solution approach. Section 5 gives numerical illustration analysis, and Section 6 concludes the paper.

Section snippets

Literature review

We review three streams of literature that are relevant for this paper. The first stream is the widely studied research on supplier selection or outsourcing, where research can be dated back to the early 1960s [20]. Supply management in SCs seeks the participation of good suppliers providing low cost and high quality. Selection of SC partners is an important decision involving many important factors. Supplier selection by its nature involves the need to trade off multiple criteria, to include

Stochastic fuzzy multi-objective programming models

Before the development of our stochastic fuzzy multi-objective programming models, we define various notations as follows:

Indices
icustomers
jsuppliers
Parameters:
nithe number of candidate suppliers desired by the ith customer
cijper unit purchase cost from supplier j by the ith customer
λijpercentage of items late from supplier j to the ith customer
βijpercentage of rejected units from supplier j
Didemand for item over planning period from the ith customer
uijumaximum amount of business for item to be

Solution approach

SFMOP (2) is not easy to be solved. But using a specific property of fuzzy set and the mean–variance utility function, we transform SFMOP (2) to QNMOP (5), which could be finally solved by regular solution approaches.

First we discuss the simplification of stochastic variable and fuzzy variable respectively. We use an assumption of normal distribution and constant absolute risk aversion to simplify the stochastically distributed utility function, which will lead to a mean–variance utility

Supply chain model

In this section we use the proposed model to evaluate a supply chain consisting of three levels: a set of ten suppliers, a core level represents the organizing, decision-making retail system and ten customers at the third level. Fig. 1 shows a diagram about the connections among suppliers, distributions, and customers. Each customer represents a demand assumed to be normal for a given period. The performance of each supplier is characterized with three quantitative variables: expected costs,

Conclusions and further consideration

We have developed a stochastic fuzzy multi-objective programming model for supply chain outsourcing risk management in presence of both random uncertainty and fuzzy uncertainty. Utility theory is proposed to treat stochastic data and fuzzy set theory is used to handle fuzzy data. An algorithm is designed to solve the proposed integrated model. We apply this new approach to model a supply chain consisting of three levels: a set of ten suppliers, a core distribution level and ten customers at the

Acknowledgement

This paper is supported by One Hundred Person Project of The Chinese Academy of Sciences, the National Natural Science Foundation of China (No. 70671039; No. 71073177; Grant No. 71110107024).

References (84)

  • P. Ekel et al.

    A general approach to solving a wide class of fuzzy optimization problems

    Fuzzy Sets and Systems

    (1998)
  • M. Fasli et al.

    Learning approaches for developing successful seller strategies in dynamic supply chain management

    Information Sciences

    (2011)
  • R. Florez-Lopez

    Strategic supplier selection in the added-value perspective: a CI approach

    Information Sciences

    (2007)
  • S.H. Ghodsypour et al.

    The total cost of logistic in supplier selection, under conditions of multiple sourcing, multiple criteria and capacity constraint

    International Journal of Production Economics

    (2001)
  • M. Goetschalckx et al.

    Modeling and design of global logistics systems: a review of integrated strategic and tactical models and design algorithms

    European Journal of Operational Research

    (2002)
  • S.H. Huang et al.

    Comprehensive and configurable metrics for supplier selection

    International Journal of Production Economics

    (2007)
  • C. Kao et al.

    A single-period inventory model with fuzzy demand

    Computers & Mathematics with Applications

    (2002)
  • O. Kulak et al.

    Fuzzy multi-attribute selection among transportation companies using axiomatic design and analytic hierarchy process

    Information Sciences

    (2005)
  • Z. Liao et al.

    A multi-objective supplier selection model under stochastic demand conditions

    International Journal of Production Economics

    (2007)
  • F.H.F. Liu et al.

    The voting analytic hierarchy process method for selecting supplier

    International Journal of Production Economics

    (2005)
  • L. Mikhailov

    Fuzzy analytical approach to partnership selection in formation of virtual enterprises

    OMEGA – International Journal of Management Science

    (2002)
  • W.L. Ng

    An efficient and simple model for multiple criteria supplier selection problem

    European Journal of Operational Research

    (2008)
  • M. Ojala et al.

    Investment decision-making in supplier networks: management of risk

    International Journal of Production Economics

    (2006)
  • R.F. Saen

    Supplier selection in the presence of both cardinal and ordinal data

    European Journal of Operational Research

    (2007)
  • S. Talluri

    A buyer–seller game model for selection and negotiation of purchasing bids

    European Journal of Operational Research

    (2002)
  • S. Talluri et al.

    Vendor evaluation with performance variability: a max–min approach

    European Journal of Operational Research

    (2003)
  • S. Talluri et al.

    Vendor performance with supply risk: a chance-constrained DEA approach

    International Journal of Production Economics

    (2006)
  • C.S. Tang

    Perspectives in supply chain risk management

    International Journal of Production Economics

    (2006)
  • D.J. Thomas et al.

    Coordinated supply chain management

    European Journal of Operational Research

    (1996)
  • J.H. Thun et al.

    An empirical analysis of supply chain risk management in the German automotive industry

    International Journal of Production Economics

    (2011)
  • S.A. Torabi et al.

    An interactive possibilistic programming approach for multiple objective supply chain master planning

    Fuzzy Sets and Systems

    (2008)
  • M. Vujosevic et al.

    EOQ formula when inventory cost is fuzzy

    International Journal Production Economics

    (1996)
  • V. Wadhwa et al.

    Vendor selection in outsourcing

    Computers and Operations Research

    (2007)
  • J. Wang et al.

    A possibilistic decision model for new product supply chain design

    European Journal of Operational Research

    (2007)
  • C.A. Weber et al.

    A multiobjective approach to vendor selection

    European Journal of Operational Research

    (1993)
  • C.A. Weber et al.

    Vendor selection criteria and methods

    European Journal of Operational Research

    (1991)
  • D. Wu

    Performance evaluation: an integrated method using data envelopment analysis and fuzzy preference relations

    European Journal of Operational Research

    (2009)
  • D. Wu et al.

    Supply chain risk, simulation, and vendor selection

    International Journal of Production Economics

    (2008)
  • H. Yu et al.

    Analyzing the evolutionary stability of the vendor-managed inventory supply chains

    Computers & Industrial Engineering

    (2009)
  • L.A. Zadeh

    Fuzzy sets

    Information and Control

    (1965)
  • M. Akarte et al.

    Web based casting supplier evaluation using analytical hierarchy process

    Journal of the Operational Research Society

    (2001)
  • R.E. Bellman et al.

    Decision-making in a fuzzy environment

    Management Science

    (1970)
  • Cited by (0)

    1

    Tel.: +1 402 472 4521; fax: +1 402 472 5855.

    View full text