Masakatsu MORI¹, Ryoji KOBAYASHI², Masaki SAMEJIMA², Norihisa KOMODA^2
1 Yokohama Research Laboratory, Hitachi, Ltd.
292, Yoshida-cho, Totsuka-ku, Yokohama, Kanagawa, 244-0817, Japan masakatsu.mori.vr@hitachi.com
² Graduate School of Information Science and Technology, Osaka University
1-5, Yamadaoka, Suita, Osaka, 565-0871, Japan
kobayashi.ryoji@ist.osaka-u.ac.jp, samejima@ist.osaka-u.ac.jp,
komoda@ist.osaka-u.ac.jp

Abstract: For the retailer on supply chain, decentralized ordering to multiple suppliers is an effective method to mitigate the risk that the retailer cannot sell the product to customers when the suppliers are down by catastrophic disasters. But decentralized ordering costs the retailer because the retailer procures products from the suppler whose procurement cost is expensive. For the retailers’ cost-benefit analysis of the decentralized ordering, we address developing the risk simulator on multi-tier supply chain to evaluate the effect of risk mitigation and the cost by decentralized ordering. In order to develop the risk simulator for the multi-tier supply chain, we combine the risk simulator for the 2-tier supply chain as a building block. In addition, when the 2-tier supply chain is combined, the risk simulator calculates propagation of the risk and the cost from a 2-tier supply chain to the others. Applying the risk simulator to the real supply chain with different parameter values, the authors confirmed that the risk simulator enables to find the relationship between the cost-benefit characteristic and the multi-tier supply chain model.

Keywords: cost-benefit analysis, conditional value at risk, decentralized ordering, multi-tier supply chain, risk simulator.

>>Full Text
CITE THIS PAPER AS:
Masakatsu MORI, Ryoji KOBAYASHI, Masaki SAMEJIMA, Norihisa KOMODA, Cost-benefit Analysis of Decentralized Ordering on Multi-tier Supply Chain by Risk Simulator, Studies in Informatics and Control, ISSN 1220-1766, vol. 23 (3), pp. 229-238, 2014. https://doi.org/10.24846/v23i3y201401

1. Introduction

Supply chain is a sequence of operations such as procurement, production, logistics and sale for supplying products from suppliers to final consumers. Retailers in the supply chain procure products from suppliers and sell products to consumers. In supply chain, there is a problem that catastrophic disasters such as earthquakes may prevent the procurement from suppliers[1][2]. The retailers suffer losses as the procurement risks if they cannot sell the products along with the consumers’ demand. So, it is necessary for the retailers to consider not only the procurement cost but also the procurement risk in procurement planning. In this paper, the authors define the risk as the possibility that the retailer cannot sell the products to customers when the catastrophic events happen.

Representative ways to reduce the procurement risk are to increase the urgent stock of the products and decentralize procurement of the product from multiple suppliers [3][4]. The decision-maker of the retailer has to consider how much cost and risk are changed by the urgent stock and decentralizing orders. Conventional researches have addressed evaluating the cost and the risk by simulation on a supply chain model with catastrophic events. The simulation methods to be used are Petri Net [5][6][7] and Monte Carlo method [8][9][10][11].

Most of the conventional researches address evaluating the effect of mitigating risks by the urgent stock [8][9]. The echelon inventory in a multi-tier supply chain can be used to mitigate the risk when the some suppliers are down by catastrophic events. However, it has been pointed out that the large amount of stocks needs much management cost of the stock. And the supplier cannot store the stock for more than several days. Only the urgent stock is not enough as a risk mitigation way.

Because of the problems only in using the urgent stock, some conventional researches address evaluating the effect of mitigating risks by decentralizing orders [10][11][12]. The simulation of the cost and the risk uses Monte Carlo method to decide the amount of the stock stochastically. But these researches are for 2-tier supply chain model, i.e. the supply chain model consists of a retailer and its suppliers. Actually, there are many multi-tier supply chains, which must be addressed as the support of decision-making. In addition, the conventional researches have not considered that the retailer and the suppliers change the procurement plan when the catastrophic disaster happens.