Application of Improved Genetic Algorithm to Traffic Equipment’s Rush-Repair in the Wartime

2012 ◽  
Vol 605-607 ◽  
pp. 49-52
Author(s):  
Geng Sheng Wu ◽  
Qi Yi Zhang

Traffic equipment’s rush-repairs in the wartime optimal assignment model was established. Combining the features of Job-shop scheduling problems, described the complexity of this problem. In order to find global optimal results efficiently, traditional GAs were improved and used for study of this problem. Though genetic algorithm, as an effective global search method, had been used in many problems, it had the disadvantages of slow convergence and poor stability in practical engineering. In order to overcome these problems, an improved genetic algorithm was proposed in terms of creation of the initial population, genetic operators, etc. At the end, the steps to solve the optimal model were put forward. With this model we had obtained ideal results. This shows that the method can offer a scientific and effective support for a decision maker in command automation of the traffic equipment’s rush-repairs in battlefield.

2021 ◽  
Vol 243 ◽  
pp. 02010
Author(s):  
Muhammad Kamal Amjad ◽  
Shahid Ikramullah Butt ◽  
Naveed Anjum

This paper presents optimization of makespan for Flexible Job Shop Scheduling Problems (FJSSP) using an Improved Genetic Algorithm integrated with Rules (IGAR). Machine assignment is done by Genetic Algorithm (GA) and operation selection is done using priority rules. Improvements in GA include a new technique of adaptive probabilities and a new forced mutation technique that positively ensures the generation of new chromosome. The scheduling part also proposed an improved scheduling rule in addition to four standard rules. The algorithm is tested against two well-known benchmark data set and results are compared with various algorithms. Comparison shows that IGAR finds known global optima in most of the cases and produces improved results as compared to other algorithms.


Author(s):  
W Wang ◽  
P Brunn

This paper presents an effective genetic algorithm (GA) for job shop sequencing and scheduling. A simple and universal gene encoding scheme for both single machine and multiple machine models and their corresponding genetic operators, selection, sequence-extracting crossover and neighbour-swap mutation are described in detail. A simple heuristic rule is adapted and embedded into the GA to avoid the production of unfeasible solutions. The results of computing experiments for a number of scheduling problems have demonstrated that the GA described in the paper is effective and efficient in terms of the quality of solution and the computing cost.


Sign in / Sign up

Export Citation Format

Share Document