Disruption Management in Airline Operations: A Solver-based Approach using Time-Space Network Optimization

TL;DR

This paper introduces AIRS, a solver-based disruption management system for airline operations that optimizes aircraft and crew recovery using a time-space network model, significantly improving efficiency and cost-effectiveness.

Contribution

The paper develops a novel MILP model for integrated aircraft-crew recovery on a time-space network, with a rapid passenger re-accommodation module, advancing airline disruption management capabilities.

Findings

AIRS reduces recovery costs compared to manual methods.

The system meets operational decision windows effectively.

It offers scalable and extensible decision support for airline operations.

Abstract

This paper presents AIRS, a day-of-operations disruption-recovery system. AIRS.ACR models integrated aircraft-crew recovery on a Time-Space Network (TSN) and solves a mixed-integer linear program (MILP) that enforces rotation continuity, crew legality, maintenance windows, slot capacities, and multi-leg integrity via flow-balance constraints; disruption-aware search-space construction and warm starts control combinatorial growth. A companion module, AIRS.PaxR, performs rapid passenger re-accommodation using greedy assignment and lightweight evolutionary search while preserving aircraft-crew feasibility. Across realistic evaluations, AIRS meets operational decision windows and reduces recovery costs relative to manual or sequential methods, providing a scalable, extensible decision-support capability for operations control centers.