Research On The Problem Of Iron And Steel Production Scheduling With Special Buffer

The scheduling of steelmaking-casting-hot rolling integrate production is a class of complex job-shop scheduling problems.In today's world,application of artificial intelligence algorithm has raised at an accelerating rate.A lot of work for the scheduling problem have been done by researchers.In our previous work,For one thing,a multiobjective optimization model of furnace plan is built.For another,a new composite evolutionary operator were proposed.However,there are some defects in these work.Based on existing results,we propose a new job-shop model for steel production.An improved evolution algorithm is presented to deal with a class of job-shop scheduling optimization problem in this paper.The main contents are as follows:Firstly,according to background knowledge of steel production,the integrated production process is described as a kind of two-stage job-shop model with buffer in this paper.Particularly,the concept of special buffer are introduced in this paper.We describe the characteristics of special buffer such as the physical structure,production patterns and scheduling constraints.Available mathematical models for integrated production is achieved through the scheduling of slab in the buffer.Secondly,the main improvements of evolution algorithm in this paper are as follows: Based on the mathematical model,the Active Schedules Decoding(ASD)and encoding approaches for production scheduling processes are proposed respectively.Owing to the disadvantages of the traditional competition selection,the improved competition selection operator is designed.In order to avoid illegal chromosome and reserve good characteristics of parent generation,Local Retention-Parents Crossover(LR-PPX)is provided.Finally,the experimental results of two kinds of scheduling problems,including the basic job-shop scheduling problem and the iron and steel production scheduling problem,was analyzed.First part is simulation experiment of benchmarks.Compared with the existing results,the proposed method can obtain the better convergences and optimum solutions.The simulation results are given to show the effectiveness of our approaches.In the second part,the results confirm earlier validity of mathematical model.The special buffer can improve the utilization rate of equipment significantly in every process.