Skip to main content
main-content
Top

2022 | Book

Optimisation of Robotic Disassembly for Remanufacturing

Authors: Prof. Dr. Yuanjun Laili, Yongjing Wang, Yilin Fang, Prof. Duc Truong Pham

Publisher: Springer International Publishing

Book Series : Springer Series in Advanced Manufacturing

share
SHARE
insite
SEARCH

About this book

This book illustrates the main characteristics, challenges and optimisation requirements of robotic disassembly. It provides a comprehensive insight on two crucial optimisation problems in the areas of robotic disassembly through a group of unified mathematical models. The online and offline optimisation of the operational sequence to dismantle a product, for example, is represented with a list of conflicting objectives and constraints. It allows the decision maker and the robots to match the situation automatically and efficiently.

To identify a generic solution under different circumstances, classical metaheuristics that can be used for the optimisation of robotic disassembly are introduced in detail. A flexible framework is then presented to implement existing metaheuristics for sequence planning and line balancing in the circumstance of robotic disassembly.

Optimisation of Robotic Disassembly for Remanufacturing provides practical case studies on typical product instances to help practitioners design efficient robotic disassembly with minimal manual operation, and offers comparisons of the state-of-the-art metaheuristics on solving the key optimisation problems. Therefore, it will be of interest to engineers, researchers, and postgraduate students in the area of remanufacturing.

Table of Contents

Frontmatter
Chapter 1. Introduction to Remanufacturing
Abstract
This chapter briefly explains remanufacturing and its motivation and introduces the main classes of optimisation problems in remanufacturing.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 2. Robotic Disassembly for Remanufacturing
Abstract
This chapter gives an overview of robotic disassembly for remanufacturing. The general concept, characteristics and enablers of robotic disassembly are reviewed. The chapter also outlines a systematic workflow for designing robotic disassembly cells based on techniques for assembly cell design.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 3. Product Representation for Disassembly Sequence Planning
Abstract
Most disassembly optimisation problems start with designing a mathematical representation that can describe component relations. This chapter discusses and compares three major groups of component relation models: the matrix-based model, graph-based model and hybrid-based model.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 4. Component and Subassembly Detection
Abstract
This chapter discusses the need for component and subassembly detection for robotic disassembly and introduces a two-pointer detection strategy to find removable components and subassemblies online during disassembly. The strategy is shown to be quick and simple to use for disassembly sequence planning and replanning. Only an interference matrix is required at the beginning. It is applicable to not only complete disassembly, but also selective and partial disassembly. The strategy solves a special problem in disassembly planning, interlocking,which may cause disassembly planning to fail.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 5. Modelling of Robotic Disassembly Sequence Planning
Abstract
This chapter discusses the prerequisites of robotic disassembly sequence planning as opposed to manual disassembly planning. Optimisation variables, objectives and constraints are summarised and modelled mathematically. Multiple disassembly operations with direction change and tool change, additional adjustment strategies, backup strategies, and human assistance are reflected in the models to deal with uncertainties in disassembly processes. The expected disassembly time and success rate of disassembly plans with uncertainties are formulated. The formulation reflects the possibility of failure in which disassembly may be terminated before all components are removed.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 6. Modelling of Robotic Disassembly Line Balancing
Abstract
This chapter examines the balancing of robotic disassembly lines. It details the modelling of the line balancing problem and presents various models developed in recent years. The chapter also discusses the use of probability distributions and fuzzy numbers to account for uncertainties in task durations.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 7. Evolutionary Optimisation for Robotic Disassembly Sequence Planning and Line Balancing
Abstract
The performance of an evolutionary algorithm in solving disassembly sequence planning or disassembly line balancing greatly depends on six parts: the evolutionary operator; encoding scheme; solution selection and update strategy; population initialisation; solution maintenance; and terminal condition. This chapter introduces classical single-objective evolutionary algorithms (SOEAs) with typical evolutionary operators, and multi-objective evolutionary algorithms (MOEAs) with typical solution selection and update strategies. The chapter also elaborates on common encoding schemes. Typical settings on algorithm initialisation, solution maintenance and terminal conditions are introduced to help engineers to design efficient evolutionary algorithms for robotic disassembly optimisation problems.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 8. Solutions for Robotic Disassembly Sequence Planning with Backup Actions

Uncertainties in the end-of-life conditions of products make disassembly difficult to robotise. A key problem is that existing industrial automation techniques use pre-determined robotic processes which cannot deal with unforeseen failed disassembly operations. This chapter models the completion rate of a disassembly plan and introduces the concept of backup actions. The aim of this strategy is to allow a disassembly robot to recover from failures automatically, and thus improve the system’s robustness. Experimental results indicate that the incorporation of automatic backup actions can triple the completion rate of a plan. The proposed computational method also outperforms other popular evolutionary approaches.

Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 9. Robotic Disassembly Sequence Re-planning

Based on the detection strategy introduced in Chapter 4, a ternary Bees Algorithm is presented to provide a re-planning solution, combining the disassembly order and the direction of not only detachable components but also the removable subassemblies. The algorithm possesses the merits of both the original Bees Algorithm and greedy search. Although it is designed for complete disassembly, the local search operator can be modified to adapt to both selective disassembly and partial disassembly. Due to its iterative character, the ternary Bees Algorithm is more suitable for small- and mediumscale products (where the number of components is fewer than 200, which is about ten times the number in the case studies considered in this chapter). With more decision variables, the decision-making time may increase. The re-planning process will therefore become inefficient.

Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 10. Solutions for Robotic Disassembly Line Balancing
Abstract
This chapter demonstrates the use of the model proposed in Chapter 6 and presents the solution of the robotic disassembly line balancing problem (DLBP) for benchmark cases using different optimisation algorithms. The best algorithms for large-scale multi-objective disassembly line balancing problems are shown to be NSGA-III and IBEA.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Chapter 11. Solutions for Mixed-Model Disassembly Line Balancing with Multi-robot Workstations
Abstract
This chapter discusses serial paced mixed-model disassembly line balancing with multi-robot workstations (MDLB-MR). The main difference between MDLB-MR and classic simple disassembly lines is the number of robots that can be allocated to each workstation. In MDLB-MR, a set of EOL products can be simultaneously disassembled and each product has its own set of precedence relations. The chapter compares the performance of different evolutionary algorithms (EAs) and concludes that the Problem-specific Bi-criterion EA (PBEA) outperforms the other EAs tested. Furthermore, a combination of non-Pareto-based EAs with the Pareto selection criterion can be effective at solving MDLB-MR problems.
Yuanjun Laili, Yongjing Wang, Yilin Fang, Duc Truong Pham
Backmatter
Metadata
Title
Optimisation of Robotic Disassembly for Remanufacturing
Authors
Prof. Dr. Yuanjun Laili
Yongjing Wang
Yilin Fang
Prof. Duc Truong Pham
Copyright Year
2022
Electronic ISBN
978-3-030-81799-2
Print ISBN
978-3-030-81798-5
DOI
https://doi.org/10.1007/978-3-030-81799-2