| Author: | Geda, Mohammed Woyeso |
| Title: | Integrated product design with consideration of remanufactured products and return uncertainties |
| Advisors: | Kwong, C. K. (ISE) Zheng, Pai (ISE) Lee, Carman (ISE) |
| Degree: | Ph.D. |
| Year: | 2021 |
| Subject: | Remanufacturing Product management Hong Kong Polytechnic University -- Dissertations |
| Department: | Department of Industrial and Systems Engineering |
| Pages: | xvi, 155 pages : color illustrations |
| Language: | English |
| Abstract: | Remanufacturing has gained growing attention as a product recovery strategy over the past few years due to stricter product take-back legislation, customers' awareness of environmental and health concerns of used product disposals, and economic reasons. Several multinational companies currently offer remanufactured products, often branded as 'refurbished,' alongside brand-new products. Nevertheless, previous studies have identified the design of products and the timing, quantity, and quality uncertainty of used product returns as major challenges of the remanufacturing industry. This research proposes an integrated methodology framework to address product design concerns and the uncertainty of used product returns from a remanufacturing perspective. Under this research framework, four core methodologies are proposed in this research for: i) the simultaneous consideration of assembly and disassembly concerns for the selection of fastening methods; ii) the development of a hierarchical optimisation model for the joint optimisation of configuration for new and remanufactured products with the consideration of upgrading decisions for used product returns; iii) the forecasting of used product returns for remanufacturing; and iv) the modelling of customers' preferences of product attributes and the estimation of market demands for brand-new and remanufacturing products under uncertainty For the i) aspect, fastening methods affect both the assembly of new/remanufactured products and the disassembly of products during remanufacturing. Therefore, assembly and disassembly concerns must be addressed simultaneously for fastening methods selections, which has been little studied previously from a remanufacturing perspective holistically. In this research, a methodology for fastening methods selection from a remanufacturing perspective known as FMSRem is proposed. The proposed FMSRem considers design factors that facilitate the assembly of new products during initial manufacturing and the disassembly/re-assembly processes during the remanufacturing of used products. A mathematical optimisation model is presented with an objective function to minimise the product assembly and disassembly cost. The genetic algorithm (GA) heuristic is proposed to solve the model. A laptop design case study is presented to demonstrate the effectiveness of the proposed methodology. Different scenarios regarding the degree of disassembly required and the volume of used product returns was run to validate the proposed methodology. The result has shown that the proposed methodology offers significant product assembly and disassembly cost savings. For the ii) aspect, a bilevel programming model is proposed for the joint optimisation of design configurations of new and remanufactured products considering specification upgrading for remanufactured products. The joint optimisation model involves two-level decision makings. The upper-level handles the configuration of new product variants to maximise the shared surplus of new product offerings. The lower-level deals with configuration and specification upgrading of remanufactured product variants to maximise the shared surplus of remanufactured product offerings. A non-linear integer bilevel programming (NLIBP) is proposed to model the hierarchical optimisation problem. A nested bilevel genetic algorithm (NBGA) is proposed to solve the NLIBP. Furthermore, a case study involving configuration design for new and remanufactured mobile phone variants is conducted to validate the proposed model. Four scenarios are investigated to examine the effects of model parameters on the optimal solutions with the simulation result given at last. For the iii) aspect, as remanufactured products are made from parts/modules recovered from used products, knowledge of the available quantity, timing, and quality of used product returns is crucial for successfully implementing remanufacturing. However, the uncertainty associated with the quantity, timing and quality of used product returns makes forecasting in remanufacturing a complex task. In this research, a distributed lag model (DLM) is proposed to forecast used products under uncertainty accurately. DLM's forecasting accuracy is primarily influenced by the lag function parameters' estimates, which has not been addressed in previous studies. To address this issue, a novel approach based on Markov Chain Monte Carlo (MCMC) and Bayesian inference is proposed, which can handle parameter estimations irrespective of the type and complexity of the lag function. A numerical case study is undertaken to demonstrate the proposed forecasting model and the parameter estimation methodology. Validation tests are conducted by comparing forecasting errors of the proposed parameter estimation approach with the maximum likelihood estimate (MLE) method. The result reveals that the proposed DLM based forecasting method can lead to an improved forecasting accuracy when the proposed MCMC based Bayesian approach is used for parameter estimation. For the iv) aspect, fuzzy regression (FR) and rating-based conjoint analysis are proposed for modelling customers' preferences for new and remanufactured product profiles. A multinomial logit model (MNL) is proposed, which uses FR and conjoint analysis results as inputs to estimate the demand for product profiles under uncertainties. A case study involving the design of new and refurbished laptop computers is further conducted to demonstrate the proposed approaches. Fuzzy utilities are determined for both the new and remanufactured product profiles, which are then used to estimate the market demands under three scenarios, including 'worst,' 'normal', and 'best' cases. It is envisioned that the design methodologies, approaches, and insights provided in this thesis can serve as a decision support tool during the early-stage product design processes. Furthermore, the forecasting methodology proposed can enable firms to manage the uncertainties associated with used product returns in a remanufacturing context. |
| Rights: | All rights reserved |
| Access: | open access |
Copyright Undertaking
As a bona fide Library user, I declare that:
- I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
- I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
- I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.
Please use this identifier to cite or link to this item:
https://theses.lib.polyu.edu.hk/handle/200/11567