One-of-a-kind production (OKP) is generally complex flexible production which is characterized by discrete and non-numerical production decomposition, product design uncertainty and changes in resource availability. The old production management and control system, theory and methods do not handle this situation well because these technologies are developed with a view to time-invariant or static production state in the traditional large batch size manufacturing companies. Considering the manufacturing project (i.e. production planning and execution process) of OKPs as the main line of research, and by clarifying the optimal control and operational management of interim product production from a viewpoint of control system, we aim at analyzing the relationship between the control system actions and the crucial production indices that characterize working time expenditure, man hour cost and workforce consumption; and developing a closed-loop dynamic production cost control and optimization system of interim products based on working hours and manpower for OKP. The research involves the following aspects:
First: A method for workforce allocation and working time optimization problem with discrete and non-numerical constraints in OKP. With a top-down refinement method to specialize the product design and production decomposition in OKP, we suggest three unit task structures. For the typical double-level-nested parallel structure, according to analysis of complex industrial scenarios, a discrete-nested-set DP (dynamic programming) is presented to solve the workforce allocation and working time optimization problem.
Second: The optimal control methods based on MLHPP (the multilevel hierarchical PERT-Petri net) are proposed to analyze dynamic interim production cost control and optimization under different workforce allocation and working time scenarios in a closed-loop control system.
Third: Entropy-weighted ANP fuzzy comprehensive evaluation of interim product production processes in OKP. OKP interim product production has the nature of multi-criteria production. To optimize the production processes influenced by complex factors, we establish an influence factor system and present a method that combines subjective and objective weights based on the entropy and an analytic network process (ANP).