Stochastic methodology for service management modeling in electric power utilities

This work proposes a methodology to modeling the probability of occurrences of emergency orders in certain regions covered by an electricity distribution utility based on the historical data, in order to promote a further pro-active routing of the commercial orders. The methodology proposed aims to reduce the average service time, which is defined as the sum of the travelling time plus the execution time. The other concern refers to the priority of this type of orders, being the emergency orders of high priority when compared to the commercial ones. By first stratifying the historical data over five week days, from Monday to Friday, one can conduct a work order forecasting related to emergency services, to further discount these amount of time from the total workday hours of all available teams. A given case study has shown how one could apply this methodology to predict emergency order.