The management and implementation of energy-thrift in hospitals

Abstract

The 1984 Auditor Genera1's report on energy~thrift in the Health Service claimed that energy~thrift programmes in hospitals were not achieving their performance targets. In order to determine the reasons for this failure, twenty hospital energy~audits were analysed. It was discovered that the two principle factors impairing the performance werez- (a) The thermal conflict between thrift measures implemented on the same thermal system, and, (b) inefficient implementation strategies. In order to investigate the influences of thermal conflict and implementation strategy on the out~comes of energy~thrift programmes,four objectives were defined:~ (i) To develop a computer model which was capable of simultaneously thermally modelling the implementation of energy-thrift measures on several hospital sites. (ii) To use the model to predict the thermal and financial out~comes of various implementation strategies. (iii) To test the sensitivity of the outrcomes to changes of unit fuel prices and capital costs, and, (iv) to compare the predicted results with those actually obtained. Data from four hospital sites and ninety thrift measures were entered into the model for analysis. It was deduced that a law of diminishing returns existed between capital investment and annual savings and that thermal conflict was responsible for an average annual financial loss of 15.5%. The financial returns on the capital invested were considerably enhanced (in one case by 3ÖØ%) by spreading the financial resources over all four sites rather than by concentrating the same capital outlay on a single site. Furthermore it was discovered that there was a level of capital investment that yielded an optimal net present value over the selected project life. The programmes were more sensitive to reductions of unit fuel prices than increases of capital costs. Most of the thrift programmes remained viable (N.B. N.P.V. > Ø using the public sector discount rate _of 5% and a project life of 5 years) after the unit fuel prices had been reduced by 5Ø% of their 1985 levels. When the results of the implemented programmes were compared with the models' predictions, the accuracy of the estimated savings ranged from a 4.5% under-estimate t 2.5% over-estimate. The software was designed to find the best-fit Ventilation rate and`base temperature for each hospital site and for each thermal zone within the hospital. When the post-implementation assessments of overall air-change rate and site base temperature were compared with those predicted, the model detected changes closely matching the predictions. â The results indicated that the technique developed in this thesis has potential as a monitoring and targeting system as well as a thermal model for predicting the out- comes of energy-thrift programmes.