Passive fire protection (PFP) has been used in the oil and gas industry for many years as a method to avoid/delay global collapse of offshore installations. However, location of PFP has normally been based on simplistic assumptions, standards, guidance, and methods that do not always consider the real response of the structure to fire. The resulting PFP schemes can be conservative, leading to unnecessary cost to the operator in terms of application and maintenance costs. More importantly, there is the potential for the PFP scheme to be insufficient for the actual fire hazards, which will increase the level of risk to the personnel onboard. Fire-induced progressive collapse is a function of the level of redundancy of a structure; it is for this reason that redundancy analyses have sometimes been used as a simplistic method to calculate the level of PFP required. However, this method does not take into account the size of the fire threat against which the PFP is designed and could lead to less-than-conservative results because it considers removing only one member of the structure at a time, without considering reduction in the strength of the surrounding members as they are also being heated by the fire. Performance-based fire-collapse analysis provides an understanding of the response of the individual members, as well as the entire structural system, to fire. Understanding the failure mechanisms, susceptibility to progressive collapse of the structure, and key members that must remain in place during an accident situation allows for the optimization of the PFP scheme, protecting only the required members while allowing for local failure of redundant members. The present paper provides a comparison between the different methods, and provides case studies that have resulted in optimum PFP schemes linked to design fires on the basis of acceptable risk levels.