A SIMULATION-ENABLED DSS FOR ALLOCATING CHECK-IN AGENTS

INFOR, Aug 2003 by Chong, Kar-Li, Grewal, Manvir, Loo, Judy, Oh, Sherry L

ABSTRACT

A decision support system (DSS) is developed from a descriptive resource simulation of Air Canada's passenger check-in process previously created in 2001 with Arena 3.0. Through time study and analysis, the impacts of 9-11 on customer arrival and service times are reflected in the updated model. The integration of a Visual Basic for Applications (VBA) front-end enables Air Canada management staff to easily enter and manipulate flight and staff schedules. What-if analysis can now be performed to understand the significance of different situations and changes Air Canada may encounter daily. A key output of the DSS is a feasible staff roster to meet the airline's internal service standards by more closely matching customer arrival patterns.

Keywords: employee scheduling; passenger check-in; simulation; decision-support systems.

RESUME

Un systeme de soutien de decision est developpe d'une simulation de ressource descriptive creee dans Arena en 2001 du processus d'enregistrement des passagers d'Air Canada. Les impacts de 9-11 sur l'arrivee de client et les delais de depannage sont refletes dans le modele. Une interface conviviale qui ete cree en Visual Basic for Applications (VBA) va permettre l'administration d'Air Canada de facilement entrer et manipuler les horaires des vols et des employes. L'analyse de ce qui peut etre (Whatif analysis) va ameliorer la comprehension d'Air Canada a propos des situations et les hasards differents qu'Air Canada peut rencontrer quotidiennement. Le resultat de la simulation est un tableau de service qui propose des horaires des employees qui satisfont les standards internes d'Air Canada et les modeles d'arrivee des passagers.

Mots-cles : horaires de personnel; enregistrement des passagers; simulation; systemes d'aide a la decision.

Recd. July 2002, Rev. June 2003, Acc. July 2003

1. INTRODUCTION

1.1 Company Background

Air Canada, Canada's largest domestic and international airline, is the 10th largest commercial airline in the world. Entering the airline industry in 1937, Air Canada has grown to provide its clients with 150 flight destinations, offering 690 scheduled flights daily and transporting 23 million scheduled and charter customers annually. In addition to its airline operations, its services include vacation packages, ground handling, and training services for other airlines. In 2001, Air Canada's total revenues were $9.61 billion, with a net loss of $1.25 billion.

Given the size and scale of Air Canada, it is essential for the company to operate with the highest level of efficiency to maximize revenue and customer goodwill. In accomplishing both of these, Air Canada can maintain its competitiveness in the industry, and sustain its image of an airline carrier dedicated to value-added customer service. Achieving this position will allow Air Canada to react swiftly when changes occur, whether internally or in their external environment.

A slowing economy and the events of September 11,2001 (9-11) have heavily impacted the airline industry. In recent months, the international market has seen many changes. A renewed alliance between United and American Airlines, the merger of Japan Airlines and Japan Air System, and government assistance are actions airlines have been taking to survive in this highly cyclical industry. Many airlines have unsuccessfully attempted to restructure; airline carriers who have ceased operations include Swissair, Sabena and Canada 3000. Competitiveness in the industry has increased as cost-cutting measures are imposed, reduced fares are used to stimulate sales, and more airlines are focusing on becoming regional carriers or low-cost discount carriers. External factors have played a large role on the operations of airlines and are anticipated to impact the industry in the future.

1.2 Previous Project

In 2001, a simulation created in Arena 3.0 (Rockwell Software, Sewickley, PA) modeled Air Canada's typical Friday check-in process at the Calgary International Airport. Before the study, it was known that the passenger arrival pattern was dependent on flight schedule, destination, and time before departure. Hence, a simulation model was preferred over queuing theory as it allowed for the use of a non-stationary Poisson arrival process. To limit the scope of the project, Friday was chosen since it met the following criteria: a variety of passenger types (business and pleasure) were represented, it is typically the busiest day of the week, and it had ample and varied Air Canada departures at all 4 major check-in locations (East, West, Transborder and Express Self-Serve Check-In).

In order to create an accurate model, extensive data collection was needed, as Air Canada did not have the required historical data. The output of the descriptive model was queue length and waiting time. After validation of the model, simulation results indicated that in the spring of 2001, 11% of passengers flying with Air Canada could be expected to wait longer than the airline's own current internal time standards of 5 minutes in Hospitality Class and 2 minutes in Executive Class. The West check-in counter proved to be most problematic as it accounted for more than 50% of the missed time standards, based on the current Friday staffing schedule.