Graph theory approach to quantify uncertainty of performance measures

• Autor(a) principal: Sousa, Sérgio
• Data de Publicação: 2015
• Outros Autores: Lopes, Isabel da Silva, Nunes, Eusébio P.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/1822/35253
• topic: Data Quality; Uncertainty; Graph theory; Performance Measures; Risk determination; Science & Technology

In this work, the performance measurement process is studied to quantify the uncertainty induced in the resulting performance measure (PM). To that end, the causes of uncertainty are identified, analysing the activities undertaken in the three following stages of the performance measurement process: design and implementation, data collection and record, and determination and analysis. A quantitative methodology based on graph theory and on the sources of uncertainty of the performance measurement process is used to calculate an uncertainty index to evaluate the level of uncertainty of a given PM or (key) performance indicator. An application example is presented. The quantification of PM uncertainty could contribute to better represent the risk associated with a given decision and also to improve the PM to increase its precision and reliability.