With more and more waste products are discarded, how to recycle them has become an urgent issue. Disassembling these discarded products is a critical step to take. With disassembly, we can maximize resource utilization and greatly save manufacturing costs. There are many influencing factors in a disassembly process. In this paper we consider the impact of disassembly tools deterioration rate on disassembly time and establish a mathematical model to minimize the disassembly time. We use the advantage actor-critic algorithm in reinforcement learning to solve this model. The correctness and superiority of the algorithm are verified by comparing with the actor-critic algorithm.

Proper disassembly operation can help increase the recovery of industrial valuable supplies and end-of-life products. To solve a disassembly line balancing problem, this work focuses on a parallel layout and proposes an intelligent optimization method to maximize disassembly profits. It first formulates a parallel multi-product disassembly line balancing problem model by taking disabled workers into account. It then designs a salp swarm algorithm with innovative encoding and decoding processes. This work finally compares the proposed algorithm with a generic algorithm. Experimental results show that the newly proposed model and algorithm can well deal with the presented problem.

In the Internet era, the continuous innovation and progress of science and technology promote the renewal of electronic and electrical products and tend to shorten their life cycle. As the recycling rate of these waste products is very low, this causes a great waste of resources. How to disassemble and recycle valuable parts is a common problem faced by the world. In essence, the recycling of waste products by enterprises is to obtain most valuable parts and components from obsolete products to gain profits. This paper considers the traditional linear disassembly line, which is widely used in factories at present. By combining the Brainstorming optimization (BO) algorithm with the K-means clustering algorithm, this work proposes a novel Improved Brainstorming optimization algorithm to obtain the near optimal solution quickly. It is compared with an Artificial Bee Colony algorithm and Gray Wolf optimization algorithm to verify its superiority in solving disassembly line balancing problems.

This article studies a steelmaking-continuous casting (SCC) scheduling problem by considering ladle allocation. It takes technological rules in steel manufacturing and ladle-related constraints into account. A scheduling problem is formulated to determine allocation equipment for jobs, production sequence for jobs processed by the same equipment, and modification operations for empty ladles after their service for jobs. To ensure the fastest production and least energy consumption, we present a mixed integer mathematical programming model with the objectives to minimize the maximum completion time, idle time penalties, and energy consumption penalties related to waiting time. To solve it, we develop a two-stage approach based on a combination of scatter search (SS) and mixed integer programming (MIP). The first stage applies an SS algorithm to determine the assignment and sequence variables for charges. For the obtained solution, we construct a temporal constraint network and establish an MIP model at the second stage. We apply ILOG.CPLEX to solve the model and find the final solution. We analyze and compare the performance of the proposed approach with a hybrid method that combines a genetic algorithm with MIP on instances constructed from a real iron-steel plant. To further verify the effectiveness of the proposed algorithm, we compare its results with optimal solutions of the constraint-relaxed original problem. The experimental results show the effectiveness of the proposed approach in solving the SCC-scheduling problem.

This paper studies a new scheduling problem in a steel plant, referring to continuous casting (CC), reheating furnace, and hot rolling (HR) processes, which is meaningful and important to the production efficiency and energy saving. First, the problem is modeled as a combination of two coupled subproblems: one assigns casts to continuous casting (CC) machines, decides sequence and start time for casts and rolling units; and another assigns furnaces and decides start time for rolling slabs in a reheating furnace. The objectives are to maximize the number of slabs processed in a mode of hot charge rolling or direct hot charge rolling so as to reduce the energy requirement and the temperature drop of slabs and minimize the residence time of slabs in a reheating furnace to save energy. Then, based on a Benders decomposition strategy, a hybrid algorithm that combines mixed-integer programming and constraint programming is designed to solve each subproblem. An effective cut-generation scheme based on a priority relationship is developed for resolving resource conflicts and unsatisfied setup time constraints. Finally, extensive experiments are conducted to verify the effectiveness of the proposed approach.

Caching at the small base station (SBS) is a promising architecture to alleviate the highly-loaded wireless video networks. SBS can cache popular video files, thus serves mobile users without going through backhaul connection to the core network and provides content- level offloading. This paper proposes a novel method for the content caching problem that optimizes cache performance. Our proposed algorithm learns the content popularity profile by predicting the probability of files to be requested, and then refreshes the cache based on the learned content popularity. Popularity learning method runs in an online fashion and has no assumption of the file requests, thus it can be used for predicting either fixed or time-varying popularity. Our simulation results show that our proposed algorithm has similar performance compared to the traditional algorithms when the content popularity profile is fixed, and performs better than other algorithms when the content popularity profile is time-varying, which is more realistic.

Distributed filtering in network is a fundamental problem in the field of network signal processing. Each node estimates or tracks some unknown state relying on the private observation and the fusion information from the network. Network fusion is generally a way of interaction over network, by which nodes can learn from each other and make decision mutually. Unlike conventional methods, we construct a distributed filter using Bayesian network game as a fusion tool, where all the nodes exchange their best strategies instead of exchanging local estimators. The proposed algorithm is a coalition of signal processing and game theory in network, which can be extended to more general signal processing and decision making models.