Improved Genetic Algorithm Researches Transportation Problem

Transportation problem is a special linear programming problem. It can solve the goods of transportation and the vehicle of dispatch reasonably. There are some problems in life, they also can be transportation problem after suitable transformation. So, transportation problem is extensive in practical life. Researching transportation problem is very meaningful.With improving science and technology, increasing productivity,many researchers research transportation problem gradually. The traditional algorithm of solving transportation problem is the tabular method. But the method is complex, it has a large amount of calculation and it is difficult to solve by computer.especially, it is not suitable for solving the larger balanced transportation problem. So, exploring new methods of solving transportation problem is a hot problem.The paper knows transportation problem and its algorithms deeply, gives the operation of genetic algorithm about solving the balanced transportation problem.based on the disadvantages of genetic algorithm, researches how to solve balanced transportation problem with improved genetic algorithm. The main work and results are the following:1. The paper introduces the background, construction and present situation of transportation problem, the description, model, component and algorithms of transportation problem, shows some problems about the algorithms.2.The paper recounts the operation of genetic algorithm carefully(include coding mode, initial population, fitness function, selection operator, crossover operator, mutation operator, crossover probability, mutation probability and stopping rules), shows the advantages and disadvantages of genetic algorithm, introduces the specific application of genetic algorithm.3. The paper gives the operation of genetic algorithm about solving the balanced transportation problem. based on the disadvantages of genetic algorithm, the paper improves disadvantages, that is, improves selection operator, crossover operator and mutation operator (special formula transformation of selection operator, uniform crossover operator, overturn mutation operator), uses adaptive crossover probability and mutation probability. The example proves that improved genetic algorithm is very superior.