Several electric utilities (viz., Virginia Power) in the United States apply demand charge on the basis of coincidental peak hour load (PHL). This is in addition to the usual kWhr charge. The hour during which the utility experiences the highest demand (MW) for the month is called the peak hour and this MW value is referred to here as the PHL. The hourly demand meter readings for the appropriate consumer (e.g., industrial or large commercial customer, wholesale purchaser, etc.) are examined to determine their demand during the peak hour of the month. The monthly demand charge is computed on the basis of the consumer's contribution to this PHL. This monthly capacity charge can run into millions of dollars for industrial and large commercial customers. Thus it is very valuable for the concerned consumer to receive advance information about the time and size of the utility's peak load. This information would enable them to take load control actions such that their demand can be reduced during the hour of peak load. The motivation behind the research reported in this paper is precisely this-to be able to predict the time and size of this PHL. The load forecast software is part of a load forecasting-load management decision support system based on microcomputers. An IBM-RT/PC is used as the central computer around which the load forecasting and load management simulator has been built in our laboratory.