Study Of Three Novel Production Planning Models And Solution Algorithms In Hybrid Production Systems Based On The Internet Of Things

With the rapid development of advanced manufacturing, the environmental problem is increasingly serious, and raw material consumption has increased dramatically, therefore, handling of waste products is gradually being valued by many countries. Manufacturers can satisfy the environmental protection requirements by properly reusing returned products for extracting their remaining value, and reduce the demand of raw materials and the production cost as well. And with the innovation of manufacturing technology, manufacturers’ ability to reuse returned products is improved continuously, and reuse ways is increasing. Thus, how to use appropriate methods to extract the remaining value from the waste products has become an important issue as a common concern of manufacturers and researchers. The key to solving this problem is to grasp the detailed information when the product left the manufacturer to enter the using stage, in order to make appropriate decisions for each waste product in the recovery process and the reuse process. The emergence of the technology of Internet of things (IoT) perfectly fills the technology of information collection in the using stage and the recovery process, which allows companies or individuals to track, monitor, and record critical real-time information anywhere via sensor networks and wireless transmission network based on a unique identification code embedded in the product. The advantage of the Internet of things technology can be fully exploited in the hybrid production system, especially keep tracking and monitoring of product in the using stage and recovery process. So how to make the Internet of things technology into actual production, to achieve the purpose of product lifecycle management, to make efficient and rational production planning decisions according to the detailed information provided by the Internet of things has become a critical issue. In this dissertation, three novel production planning problems in hybrid production systems are studied based on the Internet of things from the perspective of manufacturers in the context of electronic product manufacturers and iron and steel products manufacturing companies.This dissertation focuses on production planning problems of short life cycle products and long service life products with hybrid production system, studies the integrated planning problem of returns recovery, production and other aspects from the perspective of manufacturers in the Internet of Things. From the perspective of closed-loop supply chain, three-stage mode of recovery, hybrid production and procurement, and integrated planning with resource capacity under stochastic demand, respectively, the relevant mathematic models of production planning are established, and the structures of corresponding IoT systems are designed, then efficient algorithms are developed based on features of each problem. In this dissertation, the research results of three novel production planning models and algorithms in hybrid production systems based on IoT are summarized as follows:(1) The production planning problem in hybrid production systems based on IoT with returns value deterioration is studied, and the planning objective is to minimize total cost of the hybrid production system. An appropriate IoT framework of the hybrid production system in closed loop supply chain is developed, and then a mathematical model is established based on detailed information provided by IoT. The performance and applicability of the model is illustrated by numerical examples and sensitivity analysis. The simulation experimental results show that the presented model can effectively solve returns reusing problems with the value deterioration.(2) The production planning problem with acquisition pricing and suppliers selection in hybrid production systems based on IoT is studied, and the planning objective is to maximize total profits of the whole product life cycle. An appropriate IoT framework of the three-stage hybrid production system is developed which contains used products acquisition, hybrid production, and raw material procurement, and then a mathematical model of the three-stage hybrid production system is established based on detailed information provided by IoT. According to the characteristics of the model, a novel particle swarm optimization (PSO) based on two heuristic processes is designed to slove the problem. Then, the performance of the algorithm and the effectiveness and practicability of the model are illustrated by numerical examples and parameters sensitivity analysis.(3) The stochastic production planning problem with resource capacity planning in hybrid production systems based on IoT is studied, and the planning objective is to minimize the total expected costs of the hybrid production system. An appropriate IoT framework of the hybrid production system is developed, and then a mathematical model of the stochastic production planning integrated the resource capacity planning is established based on historical information of demand and product data provided by IoT. According to the structure characteristics of the model, a heuristic algorithm based on Lagrange relaxation is designed. Then plenty of instances are used to do the further test on the algorithm, and the test results show that the algorithm can derive a solution within one minute that the mean gap between the solution and the lower bound is 1.25%. Finally, the effect of demand uncertainty on the solution is discussed in detail by sensitivity analysis of the scenario number.