Analytic network process and multi-period goal programming integration in purchasing decisions

https://doi.org/10.1016/j.cie.2006.12.006Get rights and content

Abstract

This paper presents a multi-period inventory lot sizing scenario, where there is single product and multiple suppliers. By considering multi-period planning horizon, an integrated approach of Archimedean Goal Programming (AGP) and Analytic Network Process (ANP) is suggested. This integrated approach proposes a two-stage mathematical model to evaluate the suppliers and to determine their periodic shipment allocations given a number of tangible and intangible criteria. In the evaluation stage, the suppliers are evaluated according to 14 criteria that are involved in four control hierarchies; benefit, opportunity, cost and risk (BOCR). In the shipment stage, a multi objective mixed integer linear programming (MOMILP) model is described to solve the order allocation problem. This MOMILP model is suggested to achieve target values of periodic goals: budget, aggregate quality, total value of purchasing (TVP) and demand over the planning horizon, without exceeding vendor production capacities. This multi-period model is solved by using AGP. Finally some computational experiments are conducted to test the performance of the proposed method.

Introduction

Companies need to work with different suppliers to continue their activities. In manufacturing industries the raw materials and component parts can equal up to 70% product cost. In such circumstances the purchasing department can play a key role in cost reduction and supplier selection is one of the most important functions of purchasing management (Ghodsypour & O’Brien, 1998).

Several factors may affect a suppliers’ performance. Dickson (1966) identified 23 different criteria for vendor selection including quality, delivery, performance history, warranties, price, technical capability and financial position. Hence, supplier selection is a multi-criteria problem which includes both tangible and intangible criteria, some of which may conflict.

Basically, there are two kinds of supplier selection problem. In the first kind of supplier selection, one supplier can satisfy all the buyer’s needs (single sourcing). The management needs to make only one decision; which supplier is the best. In the second type (multiple sourcing), no supplier can satisfy all the buyer’s requirements. In such circumstances management wants to split order quantities among suppliers for a variety of reasons including creating a constant environment of competitiveness. Several methods have been proposed for single and multiple sourcing problems in the literature (Aissaoui, Haouari, & Hassini, 2006).

First publications on vendor selection can be traced back to the early 1960s. Although, the problem of supplier selection is not new, quite a few researchers treat the supplier selection issue as an optimization problem, which requires the formulation of an objective function (Wang, Huang, & Dismukes, 2004). Since, not every supplier selection criterion is quantitative, usually only a few quantitative criteria are included in the optimization formulation. To overcome this drawback, Ghodsypour and O’Brien, 1998, Wang et al., 2004 proposed integrated approaches. They are achieved in two phases. At first, a supplier evaluation is elaborated using a multi-criteria tool. In these studies, the Analytic Hierarchy Process (AHP) method was applied to make the trade-off between tangible and intangible factors and calculate a rating of suppliers. The second stage of these global approaches consists of effectively selecting the suppliers and allocating orders using mathematical programming to take into account the system constraints. Thereby in the study of Ghodsypour and O’Brien (1998), calculated ratings are applied as coefficients of an objective function in a linear program such that the total value of purchasing becomes a maximum. This single period single item model was constrained by the demand, capacity and quality requirements. They used ε-constraint method to solve the problem. In addition to a goal that maximizes the total value of purchase (AHP ratings input), Wang et al. (2004) considered a second goal that minimizes the total cost of purchase. The resulted preemptive goal programming (PGP) determines the optimal order quantity from the chosen suppliers considering as constraints vendor capacities and demand requirements. Demirtas and Ustun (2005a) combined ANP and MOMILP model to solve order allocation problem by using a Reservation Level Driven Tchebycheff Procedure. They minimized total defect rate and total cost of purchasing, and maximized TVP. On the other hand, none of these integrated approaches considered a multi-period planning horizon.

Several researchers combine supplier selection and procurement lot-sizing by considering a multi-period planning horizon and defining variables to determine the quantity purchased in each elementary period (Bender et al., 1985, Buffa and Jackson, 1983, Tempelmeier, 2002, Basnet and Leung, 2005). Buffa and Jackson (1983) presented a schedule purchase for a single product over a defined planning horizon via a GP model considering price, quality and delivery criteria. It included buyer’s specification such as material requirement and safety stock. Bender et al. (1985) studied a purchasing problem faced by IBM involving multiple products, multiple time periods, and quantity discounts (the type of quantity discount was not mentioned). The authors described, but not developed, a mixed integer optimization model, to minimize the sum of purchasing, transportation and inventory costs over the planning horizon, without exceeding vendor production capacities and various policy constraints. Contrary to single period models dealing with any form of price discount, by considering inventory management in a multi-period horizon planning, Tempelmeier (2002) incorporates a trade-off between ordering larger quantities to get a reduction on purchasing costs and maintaining low inventories to minimize holding costs. Basnet and Leung (2005) balance ordering and holding costs in a multi-item model by considering a multi-period scheduling horizon. They proposed an uncapacitated mixed linear integer programming that minimizes the aggregate purchasing, ordering and holding costs subject to demand satisfaction. The authors proposed an enumerative search algorithm and a heuristic to solve the problem. Although they considered multi-period planning horizon, neglected intangible criteria.

Both of multi-period planning horizon and intangible criteria can not be neglected in real life problems such they are working on. To eliminate this drawback, Demirtas and Ustun (2005b) have also used ANP and GP approach for multi-period lot-sizing.

In this paper, an integration of ANP and multi-period MOMILP is proposed to consider both tangible and intangible factors for choosing the best suppliers and define the optimum quantities among the selected suppliers. This two-stage mathematical model is proposed to evaluate the suppliers and to determine their shipment allocations given a number of tangible and intangible criteria.

In the evaluation stage, the suppliers are evaluated according to 14 criteria that are involved in four control hierarchies; BOCR. It will be useful to determine priorities by ANP, a new theory that extends the AHP. With the ANP it is recognized that there is feedback between the elements in different levels of the hierarchy and also between elements in the same level, so the decision elements are organized into networks of clusters and nodes. ANP deals systematically with all kinds of feedback and interactions (inner and outer dependence). When elements are linked only to elements in another cluster, the model shows only outer dependence. When elements are linked to the elements in their own cluster, there is inner dependence. Feedback can better capture the complex effects of interplay in human society (Saaty, 2001). Using pairwise comparison, a more accurate scoring method that has been applied on supplier selection are the AHP and ANP approaches. Barbarosoglu and Yazgaç, 1997, Yahya and Kingsman, 1999, Masella and Rangone, 2000, Tam and Tummala, 2001, Lee et al., 2001, Handfield et al., 2002, Colombo and Francalanci, 2004 propose the use of AHP to cope especially with determining scores. In a recent study, Liu and Hai (2005) presented the voting analytical hierarchy process (VAHP) that is a novel easier weighting procedure in place of AHP’s paired comparison. The analytical network process (ANP), a more sophisticated version of AHP, was also applied for vendor selection by Sarkis and Talluri, 2000, Sarkis and Talluri, 2002.

In the shipment stage, a MOMILP model is described for a multi-period inventory lot sizing scenario, where there are multiple suppliers and single product. It is suggested to achieve target values of periodic goals such that budget, aggregate quality, TVP and demand over the planning horizon, without exceeding supplier production capacities. This multi-period model is solved by using AGP and some computational experiments are conducted to test the performance of the proposed method. GP is an analytical approach devised to address decision-making problems where targets have been assigned to all the attributes and where the decision maker (DM) is interested in minimizing the non-achievement of the corresponding goals. In other words, the DM seeks a Simonian satisficing solution (i.e., satisfactory and sufficient) with this strategy (Romero, 2004). Buffa and Jackson, 1983, Sharma et al., 1989, Chaudhry et al., 1993, Karpak et al., 1999, Wang et al., 2004 used GP method in purchasing decisions. In these studies, only Wang et al. (2004) proposed an integrated approach for single period.

Section snippets

ANP and multi-period AGP integration in supplier selection and order allocation

As mentioned in Part 1, this integrated approach proposes a two stage mathematical model which includes supplier evaluation and shipment allocation. Fig. 1 shows the algorithm of this integrated approach. Evaluation and shipment stages are briefly summarized as follows:

A numerical example

This numerical example presents a two-stage mathematical model to select suppliers and their shipment allocations given a number of tangible and intangible criteria. In the first stage, four different suppliers that produce the plastic parts of a refrigerator plant were evaluated by using ANP. This plant is one of the leading refrigerator producers in the world with a record production of almost 2 million units in 2003. It has a wide product range with 51 main models and 900 model versions that

Conclusion

Although, supplier selection problems are not new, quite a few researchers treat the supplier selection issue as an optimization problem, which requires the formulation of the objective function. Several mathematical models have been proposed for single and multiple sourcing supplier selection problems in the literature but, not every supplier selection criterion is quantitative, so only a few quantitative criteria are included in the optimization formulation in these studies. To overcome this

References (29)

  • P.S. Bender et al.

    Improving purchasing productivity at IBM with a normative decision support system

    Interfaces

    (1985)
  • F.P. Buffa et al.

    A goal programming model for purchase planning

    Journal of Purchasing and Materials Management

    (1983)
  • A. Charnes et al.

    Optimal estimation of executive compensation by linear programming

    Management Science

    (1955)
  • A. Charnes et al.

    Management models and industrial applications of linear programming

    (1961)
  • Cited by (153)

    • Risk-averse supplier selection and order allocation in the centralized supply chains under disruption risks

      2021, Expert Systems with Applications
      Citation Excerpt :

      These category included linear programming (Anthony & Buffa, 1977; Pan, 1989), mixed-integer linear programming (Basnet & Leung, 2005; Chaudhry et al., 1993; Demirtas & Üstün, 2008; Hosseini et al., 2019), mixed-integer non-linear programming (Esmaeili-Najafabadi et al., 2019; Hu et al., 2018; Kamali et al., 2011; Keskin et al., 2010; Mendoza & Ventura, 2010; Rezaei & Davoodi, 2011; Ware et al., 2014), goal programming (Buffa & Jackson, 1983), stochastic programming (Hammami et al., 2014; Kara, 2011; Sawik, 2015, 2017, 2019), dynamic programming (Masella & Rangone, 2000; Mendoza et al., 2008), and multi-objective programming (Demirtas & Üstün, 2008; Kamali et al., 2011; Li et al., 2016; Sodenkamp et al., 2016). and 2) Qualitative models: These models involve the analytical hierarchical process (AHP) (Mendoza et al., 2008), Fuzzy-AHP (Önüt et al., 2009), weighted point method (Hu et al., 2018), analytical network process (ANP), and vendor profile analysis (Thompson, 1990), matrix approach (Gregory, 1986), vendor performance matrix approach (Soukup, 1987), TOPSIS and Fuzzy-TOPSIS (Gupta & Barua, 2017; Önüt et al., 2009), Analytical Network Process (ANP) (Bayazit, 2006; Demirtas & Ustun, 2009; Gencer & Gürpinar, 2007; Hsu & Hu, 2009; Mendoza & Ventura, 2012). As mentioned, MINLP models are a common approach in the literature SS&OA problem.

    View all citing articles on Scopus
    View full text