The steady state optimization structure for the PAFC stack in the overall hierarchical optimization and control scheme is proposed in this paper. An easy to implement, low CPU time-consuming reduced order steady state PAFC stack model, that maps the PAFC performance satisfactorily, is used as the equality constraint equation block and variable bounds are used as inequality constraints. Functional dependence of the operating variables namely hydrogen and oxygen/air flow rates, humidifier, and cell temperature on the stack power generation are simulated. Electrolyte concentration, inferentially predicted by the model aids to identify acid drying and dilution during operation. For optimization two variables namely load current and electrolyte (phosphoric acid) concentration are considered as the optimization variables; the optimized values of which are communicated as set points for gas flow rates and humidifier temperature at the advanced control level. In the present paper steady state optimization is carried out using the sequential quadratic programming (SQP) algorithm with quasi-Newton line searching to enhance convergence. Two case studies have been performed (i) economic optimization for a PAFC stack resulting in maximization of profit and (ii) optimization to achieve a time variant electrical load based on market demand. For very high demand power, the optimizer converges to the maximum possible power and restricts the system from entering into a state of operational breakdown. The inferential reduced order model based optimization scheme showed promising potential for real-life utilization of fuel cells in remote rural areas, marine, submarine, desert, mountain terrains, oceanographic applications, and transport systems, where large computational facilities are neither possible nor feasible. © 2013 American Chemical Society.