Conference Paper

Optimisation of the Beer Distribution Game with Complex Customer Demand Patterns

Dept. of Inf. Technol., Nat. Univ. of Ireland, Galway
DOI: 10.1109/CEC.2009.4983273 Conference: Proceedings of the IEEE Congress on Evolutionary Computation, CEC 2009, Trondheim, Norway, 18-21 May, 2009
Source: DBLP

ABSTRACT This paper examines a simulation of the Beer Distribution Game and a number of optimisation approaches to this game. This well known game was developed at MIT in the 1960s and has been widely used to educate graduate students and business managers on the dynamics of supply chains. This game offers a complex simulation environment involving multidimensional constrained parameters. In this research we have examined a traditional genetic algorithm approach to optimising this game, while also for the first time examining a particle swarm optimisation approach. Optimisation is used to determine the best ordering policies across an entire supply chain. This paper will present experimental results for four complex customer demand patterns. We will examine the efficacy of our optimisation approaches and analyse the implications of the results on the Beer Distribution Game. Our experimental results clearly demonstrate the advantages of both genetic algorithm and particle swarm approaches to this complex problem. We will outline a direct comparison of these results, and present a series of conclusions relating to the Beer Distribution Game.

Download full-text


Available from: Enda Howley, Sep 28, 2015
1 Follower
170 Reads
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this paper an artificial intelligence's mono-objective algorithm known as Particle Swarm Optimization (PSO) was applied. This PSO was adapted to optimize the minimum costs in the Beer Distribution Game Problem. The PSO makes an adjustment of the policy's order of the participants in the supply chain (retailer, wholesaler, distributor and manufacturer) minimizing the costs involved into the inventory holding and resulting costs from the backlogs in orders. The supply chain was tested under three dynamic consumer demand patterns: One Step, Uniform and Cyclic. The test results were compared based on the Non-Parametric Wilcoxon's Signed-Rank Test, showing which is better to use equal or individual order policy for each participant in the chain.
    Electronics, Robotics and Automotive Mechanics Conference (CERMA), 2012 IEEE Ninth; 01/2012
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Particle Swarm Optimisation (PSO) is a Swarm Intelligence based optimisation algorithm. The algorithm consists of a population of individuals that cooperate to find the global optimum of a search space. The individual particles in the swarm move according to two influences; self-cognition and social emulation. Particles wish to return to their own previous successes and also copy the behaviour of other successful particles. PSO is simple, fast and robust which makes it ideal for many optimisation problems. This thesis introduces a new variant of the PSO algorithm, PSO with Enhanced Memory Particles, where the cognitive influence on particles is enhanced by having particles remember multiple previous successes. The additional positions introduce diversity which aids exploration. To prevent this additional diversity from hindering convergence a small memory is used and Roulette Wheel Selection is used to select a single position from memory to use when calculating particles' velocities. The research shows that PSO EMP performs better than the Standard PSO in most cases and does not perform significantly worse in any case. The work presented in this thesis adds a new empirical study to the body of PSO research. It is hoped that the research will inspire new ideas and research opportunities in PSO that lead to improved PSO performance and new applications.
    08/2014, Degree: MSc, Supervisor: Enda Howley
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Particle swarm optimisation (PSO) is both a heuristic and stochastic optimisation algorithm. The purpose of these algorithms is to give approximate solutions to problems which would be otherwise too difficult to solve. The PSO algorithm optimises the problem space as a result of particles converging on the best known solution after a period of exploration. This thesis will introduce a PSO variant with Avoidance of Worst Locations (AWL). The motion of the particles in PSO AWL will be different from that of the standard PSO as a result of their ability to remember their worst locations. The particles will use this new information to improve their search of the problem space by spending less time in the worst positions of the problem space. It is found that a subtle influence from the worst location results in the optimum performance. The proposed PSO AWL has a superior performance when compared to the standard PSO and also previous implementations of worst locations. This thesis will also examine the effect of alternative neighbourhood topologies on the performance of each PSO. It is observed that the dynamic topology, which has be dubbed Gradually Increasing Directed Neighbourhoods (GIDN), further augments the performance of PSO AWL. Each of these PSO variants are then applied to the Dynamic Economic Emissions Dispatch (DEED) problem to compare their effectiveness on constrained multi objective problems. The PSO AWL performed significantly better than the standard PSO on the DEED problem with each topology. The application of this research to the DEED model demonstrates the impact of these alternative PSO approaches to real world problem domains.
    08/2015, Degree: M.Sc. Software Design & Development, Supervisor: Enda Howley