THE CONSIDERATION OF MEASUREMENT UNCERTAINTY IN FORECAST AND MAINTENANCE RELATED DECISIONS

Abstract

In the field of operations management, most decisions are seen as being based on measurement results. However, every measurement has some uncertainty and therefore managerial decisions can be incorrect. Measurement uncertainty can cause particularly large losses in the case of maintenance decisions, where decision error can cause the breakdown of the entire production or service process. Taking measurement uncertainty into consideration the losses can be decreased. The standards, technical reports and guides dealing with measurement uncertainty use probability approach and focus on the metrology aspect. They do not support a risk based decision based on the forecast of the observed characteristics. In the proposed risk based model the realized (or lost) revenue is treated along with the occurred costs in order to maximize the profit with the help of analytic calculation and simulations. This paper suggests that by treating measurement results as a time series the time of probable failure can be predicted at a given confidence level, this can support the scheduling of maintenance tasks. The uncertainty of measurement and the uncertainty of forecasting can be treated in the same model. Key words: forecast, measurement uncertainty, risk-based maintenance, stochastic process.