In making coke through carbonization process, it is possible to produce the required high quality coke in blast furnace and reduce the cost of energy as well as increase the life of coke oven if we maintain the desired coke oven temperature by means of optimal combustion management (CM). In this study, hybrid combustion diagnostic expert system (HCDES) was developed through analyzing the trend and temperature distribution of flues in the oven and combining the domain expert's knowledge. The model also considered and incorporated the characteristics and the structure of the oven. In developing the diagnostic system, a general and accessible expert language, CLIPS (C Language Integrated Production System) was used.