Optimal Double-Resource Configuration Method Of Manufacturing Systems With Production Cycle And Production Capability Constraints

Cellular flow shop workshop layout has been widely believed to be suitable for accommodating multi-type and small batch production mode. The optimal configuration of both equipments and human resources for manufacturing systems having such layouts could largely save the production cost of enterprises while ensuring the good performance of system’s capacity and lead time. In the same time, the completeness of product or assembly could be enhanced, the WIP (work-in-process) inventory could be reduced, the production cycle could be shortened, and therefore the production cost could be decreased, and the competitiveness and benefits of enterprise could be consequently enhanced.The aim of this paper is to investigate the configuration methods of dual manufacturing resources including equipments and human resources in a cellular flow shop workshop with multi-type and small batch production mode. Specifically, to minimize the investment of manufacturing system on the condition of satisfying the average production capacity constraints and production cycle time.As the average production capacity and production cycle time have close relationships to the formation of manufacturing cells, the capacity of buffering areas, and the scheduling strategy, and could not be represented as close form. The optimization of such systems is very difficult. This research adopts an approximate approach, to replace the original constraints with approximate constraints. Through establishing a simulation platform of dual resource configuration, we get simulation data and apply regression analysis model to replace original constraints and finally obtain the approximate linear integer planning model of original problem. Through training neural network, to obtain the approximate non-linear integer planning model, and then apply branch and bound algorithm to calculate the optimal solution of this model. Finally, on the basis of the obtained total amounts of equipments and human resources, this research considers the fact that different staff has different working efficiency while the working groups having master-apprentice constraints, and give the optimal configuration solution of human resources. Based on the above research ideas, the research procedure is described as follows.Firstly, toward a cellular flow shop manufacturing system with multi-type and small batch production mode, dual resources configuration problem is formulated and the corresponding mathematical model is established. Considering the difficulties of solving the model, a solution method is put forward:to replace the original constraints by approximate constraints, and get the approximate solution through calculating the optimal solution of the approximated optimization problem.Secondly, a dual resource configuration simulation platform for manufacturing systems is established. Requirement analysis is conducted first to make clear the simulation objectives and a set of experiments to be simulated. Then, the platform is designed and developed based on the typical features of the cellular flow shop manufacturing system with multi-type and small batch production mode. Simulation model is then established with concerned parameters set for the simulation. With the resource configuration mathematical model which is established through relaxing the constraints, the initial parameters of the equipments and operators in the simulating model are obtained. Then, solving experiments for the three typical features of manufacturing systems are conducted. Finally, the orthogonal experiment design and simulation are performed to provide experiment data for the problem.Thirdly, the regression-based approximate dual-resource optimal configuration problem is solved to obtain the approximate solution of the original problem. Based on the results from simulation experiments, the multiple linear regression equation of average capacity and production cycle time is obtained through gradual regression analysis. Finally, the optimal solution of the approximate optimal configuration problem, i.e., the sub-optimal solution of the original problem, is obtained through using linear integer planning to solve the mathematical model. As validated by the simulation experiments, the solution is feasible.Fourthly, the dual-resource optimal configuration problem is solved by branch and bound method. First, BP neural network is established and trained to get the network model of average production capacity and production cycle time. Then, the two original constraints which cannot be represented by close forms are replaced by the well-trained network model, to obtain approximate non-linear integer planning mathematical model. Finally, an improved branch and bound algorithm is put forward to get the optimal solution toward the partial order monotonicity of the performance indicators of the manufacturing system. As validated by the simulation experiments, the solution is feasible.Finally, the human resource scheduling problem which considers the collaboration of multiple types of staff in the cellular flow shop is investigated. First, the characteristics of the human resource planning problem in the manufacturing system of cellular flow shop are analyzed. Then, human resource model based on the working group of multi-type staff is established to consider different salaries, employment and unemployment costs, and finally obtains the configuration results through establishing dynamic planning model which reflects the specific employing and working modes of operators. Finally, the human resource model considering the collaboration of multi-type staff having different contract periods is established. Mainly, the human costs are minimized on the condition of satisfying all the human demands in different stages. On this basis, the middle and high levels of human resources in all the stages will have least fluctuations, ensuring the enterprises have a stable human team. Through establishing the human resource planning model, the optimal configuration scheme of human resource in such enterprises is given.Through the above-mentioned research works, two methods of establishing simulation-based approximate planning models are provided for the resource configuration of the cellular flow shop manufacturing system with multi-type and small batch production mode. On this basis, the human resource optimal configuration method which considers the collaboration of multiple types of staff is studied. The outputs of this research not only enrich the theories of related academic fields, but also have wide application prospects in the concerned fields.