Computer selection for institutional procurement using stochastic multicriteria acceptability analysis

Rapid technological advancements necessitate regular updates to organizations' IT infrastructures. Mass renewals of personal computers, particularly in large-scale organizations with numerous employees, require a complex decision-making process based on cost, performance, and technical criteria. During this process, determining the most suitable option from among computer models offered by different vendors is complicated by organizations' priorities and market uncertainties. In this thesis, the SMAA-2 variant of the Stochastic Multi-Criteria Acceptability Analysis (SMAA) method is used to address multi-criteria and uncertain decision-making problems encountered in mass computer purchases. This method enables the determination of the most acceptable alternatives by considering possible scenarios, even when the decision maker's criteria weights and preferences are not precisely known. The presented decision support approach helps organizations utilize their resources effectively by holistically evaluating various criteria such as technical performance, cost, and service access. In the application study, the Pro Tower 290 G9 model achieved the highest first-choice probability at 55.5% thanks to its high multi-core performance and SSD read speed. SMAA outputs, including RAI, CWV, and CF, analyze ranking trends, the underlying criteria structures, and the stability of model preference. Thus, the study contributes to data-driven, flexible, and explainable decisions regarding strategic hardware investments.