Games, Supply Chains, and Automatic Strategy Discovery Using Evolutionary Computation

Author(s):  
T. Gosling

The use of evolutionary computation is significant for the development and optimisation of strategies for dynamic and uncertain situations. This chapter introduces three cases in which evolutionary computation has already been used successfully for strategy generation in the form of work on the Iterated Prisoner’s Dilemma, Rubinstein’s alternating offers bargaining model, and the simple supply chain model. The first two of these show how evolutionary computation has been applied to extensively studied, well-known problems. The last of these demonstrates how recent statistical approaches to evolutionary computation have been applied to more complex supply chain situations that traditional game-theoretical analysis has been unable to tackle. The authors hope that the chapter will promote this approach, motivate further work in this area, and provide a guide to some of the subtleties involved in applying evolutionary computation to different problems.

2013 ◽  
Vol 7 (1) ◽  
pp. 128-135 ◽  
Author(s):  
Yoshitaka Tanimizu ◽  
◽  
Chisato Ozawa ◽  
Yusuke Shimizu ◽  
Buntaro Orita ◽  
...  

Supply chain management has been investigated for the configuring and controlling of material and information flows among different organizations. The trend has been toward even more flexible or dynamic supply chains to find suitable business partners and enter into profitable contracts. Previous studies have proposed a two-layered supply chain model consisting of two kinds of organization: clients and suppliers. This study proposes a new model representing multi-layered dynamic supply chains and a negotiation protocol in multi-layered organizations. The organizations in the middle-layers generate both orders of parts for suppliers and offers of products for clients. Production schedules in the middle-layers continue to be modified after orders are sent to suppliers. Suppliers simultaneously generate and modify sets of production schedules for individual orders to find the most profitable order of all. The effectiveness of the model and the negotiation protocol is evaluated through computational experiments.


2021 ◽  
Vol 13 (8) ◽  
pp. 4370
Author(s):  
Rithika Dulam ◽  
Kazuo Furuta ◽  
Taro Kanno

Globalization has brought not only advantages but also risks into the supply chains. One lesser studied risk is the effect of consumer behavior in crises. The recent COVID-19 pandemic has shown that the most efficient and optimized supply chains are susceptible to consumer panic buying. There is a severe need to understand the multitude of scenarios that could manifest after a catastrophe due to the change in consumer behavior so that businesses can develop a mitigation plan. The authors have developed an agent-based model that can simulate the various outcomes of a crisis using a consumer panic buying model and a supply chain model. The model quantitatively evaluates the panic purchase intention of a consumer while assessing the impact of panic buying on the supply chain. This paper introduces the implementation of the model, focusing on output analysis of the various situational settings in disaster aftermath. Preliminary study has revealed that implementing quota policy or rationing uniformly is very effective while controlling media reports or panic buying consumers can reduce consumer demand significantly.


Author(s):  
Fabiana Lucena Oliveira ◽  
Aristides da Rocha Oliveira Junior ◽  
Luiza M. Bessa Rebelo

<p class="FonteResumo">This paper discusses transport modes supporting Uncertainty Supply Chain Model (USCM) in the case of Manaus Industrial Pole (PIM), an industrial cluster in the Brazilian Amazon that hosts six hundred factories with diverse logistics and supply chain managerial strategies. USCM (Lee, 2002; Fisher, 1997)develops a dot matrix classification of the supply chains considering several attributes (e.g., agility, cost, security, responsiveness) and argues that emergent economies industrial clusters, in the effort to keep attractiveness for technological frontier firms, need to adapt supply chain strategies according to USCM attributes. The paper takes a further step, discussing which transport modes are suitable to each supply chain classified at the USCM in PIM´s case. The research´s methods covered the use of PIM´s statistical official database (secondary data), interviews with the main logistical services providers of PIM and phone survey with a sample of firms (primary data). Findings confirm the theoretical argument that different supply chains will demand different transport modes running at the same time in the same industrial cluster (Oliveira, 2009). In the case of PIM, this implies investments on port and airport infrastructure and a strategic focus on air transport mode, due to (1) short life cycle of products, (2) distance from suppliers, (3) quick response to demand and (4) the fact that even PIM´s standard products use, in average, forty per cent of air transport at inbound logistics.</p>


Author(s):  
Leila Sakli ◽  
Jean Marc Mercantini ◽  
Jean Claude Hennet

"This research concerns the formulation of models and methods for supply chains risk analysis. An ontological approach using the KOD method (Knowledge Oriented Design) has been implemented to clearly identify relationships between the concepts of supply chain, risk, vulnerability and disturbances (critical scenarios). As a result, conceptual models of supply chains facing risk situations and critical scenarios are proposed. From the resulting conceptual models and mathematical models proposed in the literature, a multi-stage supply chain model using ARIMA models incorporating the randomness of the demand has been elaborated. In order to adapt this model to scenario criticality, constraints on orders and inventories have been taken into account. Under critical disturbances on information flows (demand) and physical flows (quality of the product supplied), constraints can be reached and supply chain behaviours can evolve toward critical dynamics or even become unstable. Supply chain vulnerabilities has been assessed and discussed."


2015 ◽  
Vol 20 (6) ◽  
pp. 603-612 ◽  
Author(s):  
Andrew Potter ◽  
Denis R Towill ◽  
Martin Christopher

Purpose – The purpose of this paper is to reflect upon the impact of the original work and provide an updated model to reflect the changing environment for supply chains. In 2000, a migratory model for supply chain evolution was proposed. Design/methodology/approach – The authors start by analysing the content of the papers that have cited the original Christopher and Towill (2000) paper. The development of an updated migratory model is informed by the findings from this, and then demonstrated through a case study of the book supply chain. Findings – Despite being the major contribution, the majority of citing papers actually use other parts of the original work, and some potential reasons for this are proposed. An extra stage is added to the migratory model, reflecting a customer centric strategy. Research limitations/implications – Given that the migratory model appears under-researched, the authors identify this as an opportunity for future research and suggest that methods less common in supply chain management are used. Practical implications – The updated migratory model can be used by supply chain managers to develop appropriate supply chain strategies for their organisations, while emphasising that many of the underlying tools to enable this reflect traditional industrial engineering approaches. Originality/value – The updated migratory model represents a new contribution to understanding the evolution of supply chains.


Author(s):  
Intaher M. Ambe ◽  
Johanna A. Badenhorst-Weiss

The purpose of this article is to demonstrate the development of a supply chain model for the automotive industry that would respond to changing consumer demand. Now more than ever, businesses need to improve the efficiency of their supply chains in order to maintain a competitive advantage. The principles of lean manufacturing and just-intime (JIT) inventory control that were renowned for helping companies like Toyota, Dell and Walmart to rise to the top of their respective industries are no longer adequate. Leading companies are applying new technologies and sophisticated analytics to make their supply chains more responsive to customer demand. This challenge is driven by fierce competition, fluctuating market demand and rising customer requirements that have led to customers becoming more demanding with increased preferences. The article is based on theoretical reviews and suggests guidelines for the implementation of an automotive supply chain model for a demand-driven environment.


2021 ◽  
pp. 1-15
Author(s):  
Sudip Adak ◽  
G.S. Mahapatra

This paper develops a fuzzy two-layer supply chain for manufacturer and retailer with defective and non-defective types of products. The manufacturer produces up to a specific time, including faulty and non-defective items, and after the screening, the non-defective item sends to the retailer. The retailer’s strategy is to do the screening of items received from the manufacturer; subsequently, the perfect quality items are used to fulfill the customer’s demand, and the defective items are reworked. The retailer considers that customer demand is time and reliability dependent. The supply chain considers probabilistic deterioration for the manufacturer and retailers along with the strategies such as production rate, unit production cost, cost of idle time of manufacturer, screening, rework, etc. The optimum average profit of the integrated model is evaluated for both the cases crisp and fuzzy environments. Managerial insights and the effect of changes in the parameters’ values on the optimal inventory policy under fuzziness are presented.


Sign in / Sign up

Export Citation Format

Share Document