Optimization of crew scheduling with uniform work distribution: A new approach using genetic algorithm

• Published in: 2013 IEEE Symposium on Computational Intelligence in Production and Logistics Systems (CIPLS) pp. 61 - 69
• Main Authors: Sasikumar, R., Pai, Harikrishna, Pradeepmon, T. G.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.04.2013
• Subjects: Cost reduction; Crew scheduling; Genetic algorithm; Genetic algorithms; Job shop scheduling; performance of the firm; Schedules; Sociology; Statistics; Transportation Company; Vehicles; Work distribution
• DOI: 10.1109/CIPLS.2013.6595201

The complexity associated with transport crew scheduling problem is much more than traditional crew scheduling problem. This is mainly due to the constraints related with the transport crew scheduling, like inability to change crew in the intermediate points and to provide minimum rest time. There are several tools used for solving this problem, like heuristics, Mathematical programming, Knowledge-based approaches, Metaheuristics etc. Most of the commercial packages use a combination of Metaheuristics and linear programming, as evident from the literature survey. Among these, the use of Genetic Algorithm has got wide attention these days because of its power and flexibility. The major parameters considered in the above models are fleet utilization and cost of operation, whereas the uniformity of work distribution among the crew is not taken care of. The proposed model addresses this issue also, by considering a trade off between work distribution and cost of operation among a set of solutions generated by genetic algorithm. The studies with the proposed model show considerable improvement both in terms of cost and fleet utilization, thus boosting the overall financial performance of the firm. It also focuses in bringing out uniformity in work distribution among the crew.