An Approach To The Facility Layout Optimization Problems By Genetic Algorithm

The facility layout problem has a great impact on the entire logistics system,information flow, production capacity, production costs and production safety.Researches of facility layout problem have great significance. Based on the geneticalgorithm, in this thesis a method of facility layouts improvement including thespecific methods of mathematical modeling and resolving is proposed. What’s more,an example for the the double-line facility layout workshop is described and resolvedin this paper. The main work of this thesis is as follows:1. A comprehensive summary of facility layout problems is given in this thesisand according to the difference of the product category, the product quantity, the sizeof the workshop and the logistics mode, the characteristics of the facility layoutsproblem are introduced respectively. This thesis also discusses the corridor of theworkshop, the P/D points and other details factors that have impact on the facilitylayout.2. In the mathematical modeling, the facility layout problems are divided intosingle-line layout problems, multi-line layout problems and double-line layoutproblems and according to the characteristics of different types, differentmathematical models are established. The double-line layout mathematical modelingis invested with more efforts in than the other two for most details of the facilitylayout problems are discussed in this model. The reason of this is that this type oflayout is close to the realistic workshop and the model is simple so that the calculatingprocess can be simplified.3. When determining the objective function, except the original goal of logisticscosts, the time costs based on the elaboration of the production time in leanproduction are considered. In the meantime, the costs of equipment movement arealso considered. Finally, the three evaluation coefficients of the three sub-objectivesare determined based on the real economic value of the three sub-objectives.4. The self-adaptive genetic algorithm is applied to resolving the three types offacility layout problems. The redundancy in the encoding is reduced in order toimprove the computational efficiency. A self-adaptive mutation method is used to avoid the premature and random search. A thorough fitness function is designed toavoid the declination of the population diversity.5. With the combination of the example of R workshop, the resolving process ofthe double-line facility layout is thoroughly stated.