Abstract Load forecasting is an important problem in the operation and planning of electrical power generation, as well as in transmission and distribution networks. This paper is interested by short-term load forecasting. It deals with the development of a reliable and efficient Kernel regression model to forecast the load in the Hydro Quebec distribution network. A set of past load history comprising of weather information and load consumption is used. A non-parametric model serves to establish a relationship among past, current and future temperatures and the system loads. The paper proposes a class of flexible conditional probability models and techniques for classification and regression problems. A group of regression models is used, each one focusing on consumer classes characterising specific load behaviour. Each forecasting process has the information of the past 300 h and yields estimated loads for next 120 h. Numerical investigations show that the suggested technique is an efficient way of computing forecast statistics.