Scheduling complex flexible job shop problems

We are developing for some years scheduling softwares for chemical industry. These softwares (developed on C++) are based on heuristics combined with optimisation methods. Our aim is to design a general methodology to cope with a large class of flexible job shop problems. We are interested to deal with the following type of problem: the flexible job shop is formed of a set of machines, each of them having a “chain of operations” and connected to some others machines. These machines can have multimaintenance periods and set-up-time. The order book is a set of jobs, each of them characterised by a set of operations, which need to be processed during an uninterrupted time period of a given length on a given machine. Not all machines are able to process a particular operation of a job. Moreover, each job can have a ready-date and a due-date. Some others constraints like shared resources, waiting-stocks, delay-time between operations and human resources are to taken into account. For the optimisation procedure, metaheuristics, like simulated annealing and tabu search, are used. The paper describes the first experiments of such methodology on a real case study