Research On Dynamic Flexible Job-shop Scheduling Problem With Machine Breakdown

Intelligent production is an important part of the new generation of intelligent manufacturing.Job shop scheduling is also one of the key links for the efficient operation of intelligent production systems.Compared with traditional manufacturing job shops,the flexible job shop can effectively organize processing equipment,processing materials and production management.In the actual manufacturing process,the environment of the job shop is complex and changeable.There will be many kinds of uncertainties such as random job arrivals,machine breakdowns,order cancellations and other dynamic events.In order to improve the stability of the production process and the flexibility of the system in response to disturbances,the Dynamic Flexible Job Shop Scheduling Problem（DFJSP）with machine breakdown is studied in this paper.A dynamic scheduling model is established for different optimization objectives,and in-depth analysis and research are carried out from the rescheduling strategy to the solution algorithm.The main research contents of this paper are as follows:（1）In-depth analysis and summary of the DFJSP with machine breakdown studied in this paper.Firstly,DFJSP with machine breakdown is described in detail.Secondly,classify and summarize according to the characteristics of DFJSP.The rescheduling strategy of dynamic scheduling is introduced,and the solution method of multi-objective DFJSP is studied and analyzed.Finally,considering the specific production requirements,different performance indicators are determined,and the overall framework for solving DFJSP with machine breakdown is formed.（2）Research on DFJS with machine breakdown based on Improved Imperialist Competitive Algorithm（ICA）.First,start from the actual production situation of the intelligent workshop of modern manufacturing enterprises.The DFJSP model was established with the maximum completion time,machine energy consumption and total delay time as the objective functions.Secondly,after the machine breakdown,the event-driven dynamic scheduling strategy is used for rescheduling.And an improved ICA is proposed to solve such problems.Finally,the effectiveness of the algorithm is verified by solving the production example and comparing the experimental results of other algorithms.（3）Research on DFJS with machine breakdown based on Convolutional Neural Network（CNN）.For the DFJSP with machine breakdown,an optimization model with the objective of maximum completion time and robustness is established.In order to make the scheduling scheme have good robustness and minimize the impact of machine breakdown,a two-stage algorithm based on CNN is proposed.The first stage is to train the prediction model by CNN.The second stage is to predict the robustness of scheduling through the model trained in the first stage.First,on the basis of the above research,an improved ICA is adopted to generate training data.Then,a prediction model constructed by CNN is proposed and an alternative metric called RMn is developed to evaluate the robustness.The experimental results show that the proposed two-stage algorithm based on CNN is effective for solving this kind of DFJSP.