Complex systems carry inefficiencies within. While IT systems and automation have been used for long to remove inefficiencies in management, there has been relatively less effort to identify and remove the inefficiencies in the IT systems themselves and production support systems are no exception. In this paper, we use agent based approach to model a typical production support system and identify the inefficiencies present and corresponding impact of these inefficiencies. We focus on two particular instances of inefficiencies and impacts. First, the sub-optimal SLA compliance , wherein we detect the presence of bands where higher SLAs can be provided by the firms for little or no extra cost and hence the lower SLAs become, in essence, an inefficient point of operation. Second, the inefficient resource handling. Determining the number of re-sources (full time equivalents) to be added or removed from an ongoing production support service engagement is a complex problem and sub-optimal determination of the same impacts the firm’s performance.