Competitive Agents Implementing Parallel Tabu Searches for Job Shop Scheduling Problem with Time Lags

This paper deals with the Job Shop scheduling problem with constraints of minimum and maximum Time Lags (JSTL). This problem is an extension of the job shop scheduling problem, with additional constraints of minimum and maximum time lags existing between successive operations of the same job. In this work, we investigate Parallel Tabu Searches implemented by competitive agents for the Job Shop problem with Time Lags. The resolution model consists of an Interface Agent and a set of n competitive Job Agents, each one containing the tabu search core, generates an initial solution, executes the tabu search process and uses the NEH heuristic process for the diversification step. Good performances of the proposed model are shown through different comparisons on benchmarks based on instances of the literature.