Hotels may encounter overbooked settings with customers potentially being walked. Walk or re-accommodation costs are not homogeneous across all customer types, with costs potentially different for loyal (branded, direct) versus non/less-loyal (3rd party-intermediated) guests. Owing to cost/satisfaction impacts – hotels may need to proactively determine which guests to re-accommodate. We formulate an overbooking model with class dependent walk out costs for a hotel with two classes of reservations - loyal members with higher walk out costs, and non-members with lower walk out costs, but with each class paying the same room rate. On the morning of the stay-date, customers arrive randomly proportionate to the number of reservations by class with the hotel potentially proactively (i.e. with rooms still available) walking non-members to avoid potentially walking members. By implementing a dynamic walk out model (walk out decisions based on mix of reservations and empty rooms), optimal walkout decisions can be made to minimize total expected walk out costs. We investigate how class dependent no-show rates and walk out costs impact optimal walk out decisions and overbooking levels. We find that changes in the no-show rates for a customer class only impact the overbooking levels of the related class whereas changes in class-specific walk out costs impact all customer class overbooking levels. This thesis offers managerial insight into a proactive/strategic walk out policy for the lodging industry, aiming to achieve optimal overbooking levels.