Low carbon economy has emerged as an important task in China since the energy intensity and carbon intensity reduction targets were clearly prescribed in its recent Twelfth Five-Year Plan during 2011-2015. While the largest enterprises have undertaken initial initiative to reduce their energy consumption, small and medium-sized enterprises (SMEs) will need to share the responsibilities in meeting the nation’s targets. However, there is no established structure for helping SMEs to reduce their efficiency gap and hence the implementation of energy-saving measures in SMEs still remains patchy. Addressing this issue, this thesis seeks to understand the critical barriers faced by SMEs and aims to develop proprietary methodologies that can facilitate manufacturing SMEs to close their efficiency gap.
Process parameters optimisation is perceived to be an effective “no-cost” strategy which can be conducted by SMEs to realise energy efficiency improvement. A unique design of experiment (DOE) introduced by Dorian Shainin offers a simplistic framework to study process optimisation, but its application is not widespread and being criticised over its working principles. In order to address the inherent limitations of the Shainin’s method, it was integrated with the multivariate statistical methods and the signal-response system in the empirical study. The nature of the research aim also requires a theoretical approach to evaluate the economic performance of the energy efficiency investment. Hence, a spreadsheet-based decision support system (file SERP.xlsm) was created via dynamic programming (DP) method.
The main contributions of this thesis can be subdivided into empirical level and theoretical level. At the empirical level, a technically feasible yet economically viable approach called “multi-response dynamic Shainin DOE” was developed. An empirical study on the injection moulding process was presented to examine the validity of this novel integrated methodology. The emphasis has been on the integration of multivariate techniques and signal-response analysis. The former successfully identified the critical factors to energy consumption and moulded parts’ impact performance regardless of the great fluctuation in the impact response. The latter enables the end-user to achieve different performance output based on the particular intent. At the theoretical level, the “DP-based spreadsheet solution” provides a convenient platform to help the rationally-behaved decision makers evaluate the energy efficiency investments. A simple hypothetical case study on the injection moulding industry was illustrated how the decision-making process for equipment replacement can evolve over time.
To sum up, both proprietary methodologies enhance the dynamicity in the optimisation process to support injection moulding industry in closing their efficiency gap. The study at the empirical level was limited by the absence of real industrial case study where it is important to justify the practicality of the proposed methodology. Regarding the theoretical level, the dataset for initial validation on the spreadsheet solution was not available. Finally, it is important to continue the future work on the research limitations in order to increase the cogency of the proprietary methodologies for common use in the industry.
|Date of Award||8 Jul 2015|
- Univerisity of Nottingham
|Supervisor||Hui Leng Choo (Supervisor) & Christopher Rudd (Supervisor)|