This paper presents deterioration models for maintenance planning of offshore jacket platform based on two methods: i) stochastic Markov-chain based model and ii) stochastic-mechanistic deterioration models based on steel corrosion rates. Markov-chain models require the estimation of transition probability matrix (TPM), which is typically derived from the inspection data. The global structural health condition of the jacket is computed based on the condition of individual elements and their criticality in terms of failure consequence. The criticality factors are established based on nonlinear static redundancy analyses. This method can model deterioration when routine inspection records of jacket members are available. When there is scarcity of inspection records, stochastic-mechanistic deterioration modeling approach can be used. Monte-Carlo simulations with established corrosion wastage models are utilized to estimate the time-dependent deterioration of jacket legs, horizontal and diagonal bracings in splash and immersion zones. This method is proposed when there is scarcity of inspection records. The deterioration models are further utilized to predict the timing for Maintenance, Repair and Rehabilitation (MRR) actions, and estimate the residual service life of the jacket platform. This study demonstrates the application of the proposed deterioration modeling approaches with a case study of a typical 4-legged offshore jacket platform.