Adiabatic oxidation techniques have traditionally been used in the Department of Mineral Resources Engineering at the University of Nottingham to investigate the problem of spontaneous combustion. Recently a computerised approach has been developed to perform real time data acquisition and control for the system. The adiabatic apparatus was interfaced with a high performance IBM personal computer (PC). A dynamic proportional temperature control strategy has been implemented to overcome the problems of continuously maintaining adiabatic conditions within the calorimeter. An optional add-on expert system has been designed as an intelligent front end to the apparatus to provide better supervision and monitoring of the test. Preliminary experiments indicated that the system performed well under different experimental conditions and adiabaticity achieved within acceptable range. The computerised system allows the experimental results to be accessed directly by an expert system through interrogating the output file from the data acquisition and control software. The information is used with the geological and mining data to provide a more comprehensive risk prediction. The risk assessment can be presented as a risk certainty factor for a specific location, or on a contour map for a mining area. This allows the engineers to focus their attention on the areas with high risk and therefore reduce the cost of preventative actions which could be overdone in areas with low risk of spontaneous heating. Further improvements of the system are identified with highlight on the development of A Real Time Intelligent Mine Fire Detection and Monitoring System. The paper concludes by stressing the need of the developed system for the study of expected increase of spontaneous combustion risk in deep underground mining and large tonnages of coal storage and shipment.