June 1997 • Volume 24 • Number 3



On Schedule to Succeed


Close partnership with Sabre pays off for SNCF, the French national railroad

By Nancy Bistritz

This is another in a series of articles based on interviews with recent Franz Edelman Award finalists. Geared toward practitioners, thearticles strive to provide lessons the Edelman authors learned and some pitfalls theyencountered during the course of their work. Persons interested in more information on this or other recent Edelman work are encouraged to read the annual special issue (January/February) of Interfaces in which the full papers, submitted in conjunction with the Edelman competition, are published. To obtain a copy of the most recent special issue of Interfaces or for subscription information to the journal, call 1-800-4INFORMs.


If you ask William Stripling, director of Sabre Logistics Group in the Dallas-Fort Worth area, what the atmosphere is like at Sabre Decision Technologies these days, he might tell you that there is an overwhelming feeling of déjà vu. That's because for the second time in less than a decade, the company's efforts have contributed to the earning of the Franz Edelman Award.

This year's Edelman Award recipient, the Societe Nationale des Chemins de Fer Francais (SNCF), in conjunction with SDT, developed two new products that have revolutionized the railroad industry and its ability to serve a rapidly growing market.


The Beginning
In the late 1980s, with the onset of deregulation and the likely introduction of more high-speed train networks on the horizon, the French national railroad (SNCF) was faced with meeting a new challenge: coming up with a system that could offer a competitive product.

"As the environment is evolving very quickly in Europe with deregulation and with the Rome market (which is very strong), SNCF has to redo its schedules quite often in order to be competitive compared to the other modes of transportation," said Cécile Queille, director of SNCF. "Before it took us a long time to reach a good product that would meet the customer demand."

A "good product" in this case is an effective means of transportation that offers the most efficient schedule at the most attractive cost. Recognizing the need for a well-designed system, SNCF sought assistance in devising a new system that could attract customers and increase revenue while decreasing operating costs.

Designing such a system was no easy task, considering that SNCF is the third largest railroad in the world. SNCF was the first railroad to introduce a high-speed network in 1981. Its electric powered Trains a Grande Vitesse (TGV) carries more than 50 million passengers per year to 140 European cities. In addition to the TGV, the SNCF deploys more than 300 train units and handles more than 3,000 scheduled departures per week.

In 1987, the SNCF began searching for a larger system that could handle its customer demands as well as the demands of the future while at the same time providing better inventory management. The result was a partnership with Sabre Decision Technologies, a developer of transportation decision systems.

Sabre previously developed a yield management system for American Airlines which enabled the airline to increase its revenue by $1.4 billion in three years. Yield management allows any hospitality provider (i.e. hotels, airlines and rental car agencies) to overbook for no-shows.

"SNCF and Sabre Decision Technologies both felt the airline technology and the airline approach to reservations, yield management and scheduling were very similar, so we built products and systems parallel to those that are now working well in the airline industry," Stripling said.

After the implementation of a yield management system in 1989, Sabre began work on improving the scheduling system and train routes for maintenance purposes. Sabre also set out to build a system that would allow SNCF to add additional capacity to trains a few days prior to their departure if reservation activity was high.


Implementation
"In 1990, we conducted two proof-of-concept studies with the SNCF — taking their data, working side by side with their operations people and in a sense proving to them that the models we intended to build could indeed both save them money as far as cutting costs but could also increase revenue," Stripling said.

The SDT/SNCF joint work produced two new products: RailCap, which allows SNCF to add additional reservation capacity, and RailPlus, which enables the railroad to build new schedules.

In an effort to determine effectiveness, both products were used in parallel with the old method at the SNCF. That is, users conducted business as they always had, while simultaneously employing the new systems. According to Queille, this allowed users to witness, firsthand, what kinds of effects the new system would have.

"In a similar manner, we could enhance the prototype by using it," she said. "By involving the future users in the design and the iteration, it greatly helped the organization to accept everything."

However, not all users were comfortable with the system. Both SDT and the SNCF admit that there was much skepticism at the onset of the project. Not only were the users being introduced to new technology, they were also incorporating two separate business functions into one task. One of the benefits of RailPlus and RailCap is that they merge technology with customer satisfaction. Unfortunately, employees in the marketing field and those working in production were not used to a system that simultaneously catered to both fields.

"Marketing people seemed interested in this system very quickly, but the production people were very doubtful," said Queille. "They thought they were doing a great job by hand — that no one could do it better. We had to prove to them that the system was doing as good as they were doing or better."

And that it did. Prior to using RailCap and RailPlus, reconfiguring a schedule could take as long as two to three years. Using the new systems, a schedule can be redone in a matter of months.

According to Queille, employees were nervous that the system would cost some of them their jobs, but that didn't turn out to be the case. Instead, two years after the system's inception, no one has been laid off, but the employees are performing dramatically different jobs.

"(The implementation) changed the concept of the work," she said. "But we have not decreased the overall number of people working because now some people are using RailPlus to do prospective and strategic studies to see what kind of investment we should do in four to five years time. It is a new way of working, it is a new way of thinking, and that has greatly enhanced the content of the work."

With a new way of doing business came the introduction of an operations research department within the SNCF.

Nejib Ben-Kheder, director in charge of passenger surface transportation solutions at SDT, said the development of an OR division has both SDT and the SNCF literally working hand-in-hand.

"SNCF eventually saw that operations research is a strategic activity, and they asked us to help them put into place an operations research department. This was an outcome of this project that was not planned," Ben-Kheder said, adding that c the operations research department at the SNCF has about 30 people working in it — 15 from the SNCF and 15 from SDT.


Results
Both systems are enabling the users at SNCF to think and work like never before, leading to more challenging work with more comprehensive results.

"Prior to using RailCap, an analyst would assign an extra train section on a service that showed high reservation activity, but they wouldn't take into account the cost of sending that train section out to wherever it had to go," said Stripling. "RailCap takes into account both the extra revenue you're going to receive from that extra train section, and it also considers how much it's going to cost you to put that extra train section into service."

Very quickly after the implementation of the two systems, the SNCF saw results — most importantly a dramatic change in the way the SNCF was building its schedules.

"There were scenarios that they were not able to run before; the data was not available. Enhanced productivity was something that struck everybody. That was something we saw from Day One," said Ben-Kheder.

RailCap also reduced the number of empty shuttles that were run on a daily basis.

According to Queille, because of the parallel method in which the system was used next to the older method, the SNCF was able to measure the reduction in operating costs and revenue enhancements.

"For the same schedule, without changing anything, we could reduce the operating costs by about 2 percent," she said, adding that with a reduction in operating costs came a better utilization of the SNCF's resources including trains, drivers and other personnel within the organization.

As for revenue enhancement, a more efficient train schedule brought more customers. More customers have generated greater revenue. With the demand for efficient travel constantly growing, SNCF has continually expanded its system. "We are building a new high-speed track toward the Riviera which is going to be implemented around the year 2000," said Queille.

"We are using RailPlus to set up the best schedule and to set up the optimal size fleet."


Lessons Learned
Since the development and implementation of both systems, SDT and SNCF have both been approached by other railroads in Europe. Ben-Kheder reports that SDT has talked with both the Italian and German railroads. "I think they're quite fascinated by what RailPlus does," Ben-Kheder said.

Throughout the five-year process, SDT worked side-by-side with the end-users at SNCF, allowing both parties to gain a better understanding of the products and to make changes. "We worked with the client early on trying to gain their support," said Ben-Kheder. "So, when we put these systems into production, it was a fairly smooth process. This is something we've been preaching at SDT."

The SDT/SNCF collaboration also proved that success can travel from one mode of transportation to another. "We were successful in migrating this technology from the airline industry to passenger rail," Stripling said, "just as we were successful in migrating yield management from airlines into the cruise ship, hotel and rental car industries."
Nancy Bistritz is the assistant editor of
OR/MS Today.


For more information, put the number 1 in the appropriate space on the
Reader Service Form



E-mail to the Editorial Department of OR/MS Today: orms@lionhrtpub.com


OR/MS Today copyright © 1997, 1998 by the Institute for Operations Research and the Management Sciences. All rights reserved.


Lionheart Publishing, Inc.
2555 Cumberland Parkway, Suite 299, Atlanta, GA 30339 USA
Phone: 770-431-0867 | Fax: 770-432-6969
E-mail: lpi@lionhrtpub.com


Web Site © Copyright 1997, 1998 by Lionheart Publishing, Inc. All rights reserved.
Web Design by Premier Web Designs, e-mail lionwebmaster@preweb.com