Center for the Built Environment

This report describes the results of a study designed to assess the accuracy of the complaint prediction model proposed by Federspiel (2000), and to re-calibrate it if necessary. The complaint prediction model predicts the average number of hot and/or cold complaints per square foot per year as a function of the statistical behavior of the indoor temperature. The significance of the model is that it allows us to quantitatively assign economic cost to thermal discomfort in buildings because the operations and maintenance labor cost associated with handling thermal complaints can be accurately quantified.

We collected temperature time series data from six buildings ranging in size from 60,000 square feet to 800,000 square feet. They were located in the Seattle, San Francisco, and Minneapolis areas. In two of the buildings we collected temperature time series twice, once during the winter and once during the summer. In the other four we collected the temperature time series just once. All six buildings used a computerized maintenance management system (CMMS). We determined the number of hot and cold complaints during each temperature monitoring interval from the CMMS data.