In general, making optimal decisions is a never ending challenge that decision makers face. A comprehensive model that integrates decisions at all three levels of decision making (i.e., strategic, tactical, and operational) can help the decision maker to find solutions that best serve the organization's performance. However, as organizations expand their business, more decision problems that are larger in size and more complicated need to be considered. Thus, in practice such comprehensive models become very difficult to develop and solve. The challenge becomes even more complicated as uncertainty might have a significant effect on operations and has to be taken into account explicitly. To overcome these challenges, decision makers usually focus on individual decisions separately. They make strategic decisions first and pass the outcome to the lower levels as input. This approach fails to fully acknowledge the existing dependencies between decisions. Dealing with each decision individually can lead the decision maker to prefer an optimal solution that serves the individual planning problem but overlooks the effect of that decision all the way down to the lowest levels of operation.
In this research, we propose a Reliable Integrated Planning Framework (RIPF) as a modeling approach for highly complex systems that utilizes corrective constraints to capture interdependencies between decisions at different levels of decision making. We apply this modeling approach to a simplified airline operations framework which involves planning problems associated with hub network design at the strategic level, flight times and gate selection and ground crew determination at the tactical level, and ground crew task assignment at the operational level. To apply the RIPF in this context, we develop modeling and solution methods for the Reliable p-Hub Network Design Problem under Multiple Disruptions (RpHND-MD) for the strategic level problem, and the Task Assignment Problem with Flexible Execution Times and Sequence Dependent Travel (TAP-FET-SDT) for the operational level problem. Extensive computational testing of the RpHND-MD and the TAP-FET-SDT problems is completed to evaluate the performance of the modeling and solution methods developed in each case and to obtain insights about these problems as different parameters are modified. Then, we formally introduce RIPF and provide details about its application in a case study of airline operations planning. We implement a simulation of failure scenarios (i.e., hub airport disruptions) affecting airline operations to compare the performance of decisions selected using a classical planning approach against decisions made using the proposed RIPF. Based on the outcomes of the comparison, it is shown that the addition of corrective constraints to capture top-to-bottom and bottom-to-top dependencies between different planning levels helps to improve the overall performance of the system in terms of fewer resource shortages (i.e., gate shortages and ground crew shortages) at active airport hubs when other hubs fail in the network. At the same time, these results suggest that the RIPF provides decision makers a systematic approach to incorporate valuable information at the tactical and operational levels when making decisions in uncertain environments even when still making use of deterministic models at these levels. At the end, we discuss some of the challenges involved in applying the RIPF and conclude by suggesting directions for future research.