Loop plant modeling: Statistical analyses of costs in loop plant operations

The Serving Area Concept (SAC) involves a new procedure for the design and administration of the loop plant to reduce operating costs. Two major problems facing a loop plant engineer considering conversion to SAC are determining which areas should be converted (and in what order) and assessing the savings resulting from the conversion. This paper presents methodology and data analysis results useful for solving such problems. The data analyzed are from the Prototype District and measure a large number of facility related problems both before and after conversion to SAC. A cost penalty measure, based on observed facility problems, is calculated for a given area using data collected in that area over a certain period of time. The before conversion data are characterized and modeled in order to quantify the uncertainty, in the form of a confidence interval, associated with this cost penalty. Confidence intervals are useful to decide appropriate sizes for the data collection areas, appropriate lengths of time for data collection, as well as for comparing the results between two or more areas. The effect of conversion to SAC on the cost penalty measure is also examined. It is found that after conversion costs are much lower than before conversion costs, but that costs continue to decrease for at least 9 to 12 months after conversion takes place. The analysis and results presented here yield methods and guidelines to be used for data collection and analysis in other districts. These can help in reliably choosing areas for conversion to SAC which will maximize savings.