Font Size: a A A

Research On WIP Control Technologies In The Production Logistics Control System

Posted on:2006-05-24Degree:DoctorType:Dissertation
Country:ChinaCandidate:C F GuoFull Text:PDF
GTID:1116360152989410Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
The four target criteria of lead time, scheduling deviation, work-in-progress (WIP), and utilization determine the objectives of production logistics. As demand diversified and with very stringent requirements for product availability, it is getting increasingly important to maintain lower and steadier WIP in the value chain. As the base-control facility, the production logistics control system (PLCS) is the key factor to control WIP. Aiming at the disadvantages and drawbacks of the traditional PLCS in the dynamic environments, based on control theories and through integration of proportional & integral control (PI) and statistical process control (SPC), an integrating PLCS (I-PLCS) is proposed. An extensive study on its WIP controlling technologies is carried out. Taking unfinished parts, that is, parts' WIP, as the background, based on the continuous time differential equations, a WIP control model of integrating the feed-forward and feedback control loops in serial production lines is proposed, in which three control strategies, the demand smoothness, the production delays and the inventory replenishment rules, are sufficiently presented. Effects of system parameters on WIP are analyzed through the simulation of demand fluctuations. The PI control algorithm is proposed to apply for WIP control in serial production lines, which is widely used in process industries. Effects of different deployments of PI controllers are analyzed, and the cascade control of PI controllers is suggested. Then the design of PI controllers is transformed into an optimization problem on the frequency domain, so as to guarantee the stability and robustness of PI controllers' parameters. A case study is carried out in the production line of autobike engines' crankcases, whose simulation results confirm that the PI controller, along with the characteristic of being much robust, have better control effects than the proportional controller. For the implementation of PI controllers, the convex optimization is utilized to realize the load global optimal allocations. Effective measures are then taken to resolve the situations where WIP is beyond the limitations, that is, two kinds of controllers, the backlog controller and the input controller, are devised to stabilize WIP, which correspond with the I controller and the P controller in the PI algorithm respectively. The implementing effects of PI controllers are very convincing. The SPC/PI integration control algorithm is also being used for WIP control in serial production lines, so as to improve the performance permanently. After the introduction of how SPC keeps process under control and various statistical control charts, limitations of traditional SPC is evaluated, together with the requirements of SPC/PI integration. An SPC/PI integration model is proposed to provide the status of WIP in a serial production line, which is verified by simulation under the drift and shift upsets in the process mean. The conclusion is reached that the excellent SPC/PI integration does a better job than a single PI controller and it turns out to be a priority under CIMS environments. Based on the above researches, the implementation of the I-PLCS system is carried out in the production line of autobike engines' crankcases. Subjects such as its components, functions and I/O relations, the pricinble of SPC/PI controllers, are evaluated. Then the implementation effects of the I-PLCS system are assessed by simulating the practical scheduling scenes under dynamic environments.
Keywords/Search Tags:Work-in-progress control, Proportional and integral control algorithm, Statistical process control, integration, dynamic scheduling
PDF Full Text Request
Related items