Research On Task Decomposition,Sequencing And Allocation Optimization Of Complex Product Design Process

The competition between enterprises with economic growth is more and more intense, the competition of enterprises is the competition of the products, how to develop competitive products and short the development time and low the cost become the business focus. With the development of science and technology, the structure and function of the product is becoming more and more complex, resulting in the product design process more and more complex and having many uncertain factors. Adopting the rational management of science and technology planning process of product design is essential for enterprise survival. In this paper, the task decomposition, task sequencing, task allocation modeling and optimization techniques for complex product design process are studied.According to the task decomposition, a decomposition method based on product design phase and product structure tree is proposed. Firstly, the task encoding rules and the task decomposition rules based on the product phase and the product structure tree are presented.Based on the decomposition of the task tree, this paper proposes a hierarchical design structure matrix modeling method, any hierarchical task model can be obtained by that model method.Aiming at the difficult of calculating the time and the cost of product development process(PDP), a method of simulation of PDP based on activity overlap was proposed.Information input matrix and information output matrix were adopted to measure the situation of overlaps which was divided to forward overlap and feedback overlap; The iterative features were described by rework probability matrix, rework impact matrix, the learning curve and quality improvement matrix. Rework time-cost-quality calculation method was proposed based on activity overlap and irritation. Then an improved simulation algorithm of PDP was proposed. Finally, an example is used to illustrate the efficiency ofthe method.Aiming at the task sequencing, a method of task sequencing optimization based on genetic algorithm and simulation is proposed. First of all, the shortest time to complete the process and highest quality was proposed as task sequencing's objective function with the limited cost; and then the genetic algorithm is called simulation procedures, simulation to get the target value, to solve the optimization of task scheduling.Aiming at task assignment, a task assignment model based on task iteration and feedback is proposed. Firstly, the time, cost and quality of a single task under overlapped and iterative feedback are obtained by simulation. Then, designer capability was modeled by personnel efficiency matrix, personnel cost matrix and personnel quality. Tasks were modeled by design structure matrix, workload matrix and task quality weighting matrix. Based on the matrix model, time-cost-quality task assignment model and concurrent task assignment strategy were proposed. Finally ant colony algorithm was used to allocate tasks, and an example is used to illustrate the efficiency of the method.