A hierarchical optimization structure is proposed for PAFC Stack. The performance of a PAFC stack is mapped with an easy to implement, low CPU usage consuming reduced order model, used as constraint equation block for the optimizer. The model is validated using different sets of experimental polarization data. Effect of variation of operating variables like hydrogen flow rate, humidifier and cell temperature on the load current versus voltage and power profiles are simulated. Finally, an economic optimization of a PAFC stack is carried out to get the optimum set of operating variables. © 2012 Elsevier B.V.