Abstract:
Performance-Based Contracting (PBC), e.g. Contracting for Availability (CfA), has been extensively applied in many industry sectors such as defence, aerospace and railway. Under PBC, complex support activities (e.g. maintenance, training, etc.) are outsourced, under mid to long term contracting arrangements, to maintain certain level of systems’ performance (e.g. availability). However, building robust cost and availability estimates is particularly challenging at the bidding stage because therei is lack of methods and limited availability of data for analysis.
Driven by this contextual challenge this PhD aims to develop a process to simulate and optimise cost and availability estimates at the bidding stage of CfA. The research methodology follows a human-centred design approach, focusing on the end-user stakeholders. An interaction with seven manufacturing organisations involved in the bidding process of CfA enabled to identify the state-of-practice and the industry needs, and a review of literature in PBC and cost estimation enabled to identify the research gaps.
A simulation model for cost and availability trade-off and estimation (CATECAB) has been developed, to support cost engineers during the bidding preparation. Also, a multi-objective genetic algorithm (EMOGA) has been developed to combine with the CATECAB and build a cost and availability estimation and optimisation model (CAEOCAB). Techniques such as Monte-Carlo simulation, bootstrapping resampling, multi-regression analysis and genetic algorithms have been applied. This model is able to estimate the optimal investment in the attributes that impact the availability of the systems, according to total contract cost, availability and duration targets.
The validation of the models is performed by means of four case studies with twenty-one CfA scenarios, in the maritime and air domains. The outcomes indicate a representable accuracy for the estimates produced by the models, which has been considered suitable for the early stages of the bidding process.
