The Added Value of Asset Management in Power Generation Plants in Developing Countries Context: a Case for Kenya

This dissertation explores and presents a framework for asset management of power generation facilities with a specific focus of managing such facilities in the developing countries context. The framework leverages on risk management phases proposed in the ISO 31010, and importantly, the phases propose a structured approach through which failure risks in assets are systematically identified, analyzed and consequently mitigated through formulating appropriate maintenance strategies. However, prior to implementing a risk management approach, understanding the risk concept in the context of maintenance decision making is important. From literature, however, such understanding is not straightforward which also influences the process of assessing asset failure risks. This is evident in Chapter 2 of this thesis dissertation where maintenance decision support structures so far developed seldom support robust assessment of failure risks in operable assets. Furthermore, from Chapter 2, several risk assessment methodologies are presented in asset management literature, yet a decision framework for selecting appropriate techniques while taking into account the business and operation context of the organization are lacking. This aspect is addressed in Chapter 3 where a selection methodology is developed, and based on the Analytic Network Process approach. The selection methodology takes into account practical decision making competencies necessary for performing risk assessment in the context of maintenance decision making. Apart from selection challenges, it was also apparent from literature that existing risk assessment methods were seldom dynamic in the sense that the risk assessment exercise rarely evolved with availability of new sources of risk. This aspect is addressed in Chapter 4 where a dynamic risk assessment methodology specific for maintenance decision making is developed. The methodology leverages on the hierarchical Bayes approach for updating risk assessment results with availability of new sources of risks. The practical perspective and applicability of the methodology is demonstrated through the application case of thermal power plant equipment failures. Chapters 5 and 6 evaluate a very important aspect in maintenance decision making – root cause analysis. From literature, the root cause analysis exercise is dominated by qualitative and semi-quantitative approaches which often are susceptible to decision makers’ bias. Yet, understanding the focal root causes of failure is important for developing effective risk mitigation strategies, or maintenance strategies. Thus in Chapter 5, a data exploration framework which leverages on maintenance data is developed. Importantly, the framework incorporates multivariate and cluster analysis techniques for exploring causal failure associations embedded in maintenance data. On the basis of the explored causal associations, causal maps are generated and used as the basis of identifying focal root causes of recurrent equipment failure. Chapter 6 extends the data exploration approach by incorporating association rule mining framework for extracting important causal association rules which are important for mapping the focal root causes of equipment failure. Chapter 7 extends the risk mitigation phase and explores human aspects which are responsible for maintenance errors, and embedded in maintenance data. Thus leveraging on the Reason’s model, error-provoking factors embedded in maintenance data is evaluated from which, managerial strategies are proposed for mitigating such factors. Chapter 8 presents a simulation modelling framework developed for quantifying the impact of alternative maintenance intervention strategies on power plant performance and spare part replenishment. The modelling framework considers a multi-echelon replenishment network, and further evaluates the added value of component recondition on part availability. Finally, Chapters 9 and 10 evaluate performance management aspects in maintenance decision making where a maintenance maturity model is developed for assessing maintenance capabilities within the organization. The performance management aspect also extends to a developed Hoshin-Kanri framework for deploying maintenance objectives to all organization levels – strategic, tactical and operational. Lastly, Chapter 11 presents the conclusions and contributions to the state of the art, as well as proposals for future research in the context of asset management, with focus of risk management.nrpages: 327status: publishe