Start small, think big

GAF Materials Corp., based in Wayne, N.J., ships more than nine billion pounds of roofing materials a year.

By Thomas Foster -- Logistics Management, 12/1/1998

GAF Materials Corp., the largest manufacturer of asphalt-based roofing materials in the United States, is about to take the plunge into supply chain optimization. But before it makes this leap, the company has decided to test the waters first, trying optimization on a smaller scale.

Its first venture into this area, therefore, has been a freight-management decision-support system that the company developed on its own. "With 200,000 shipments each year totaling nine billion pounds of finished product, transportation costs are a huge part of GAF's supply chain costs," says Gregory L. Schlegel, the company's supply chain manager and a past president of APICS, an organization of supply chain professionals. "That is one reason why transportation was selected as a pilot for using an optimization tool."

The purpose of the freight-optimization model is to find the least-cost dispatching option for each day's shipments. At this point, the model focuses solely on truckload shipments, which make up 70 percent of GAF's outbound traffic. (The remaining 30 percent of shipments go by rail or intermodal transport.) With 13,000 shipments moving each month from 30 plants and 10 warehouses to distributors, large-volume home centers, and job sites all over the country, the optimization task is more than a small experiment.

The data to be entered into the freight-management model include open orders for each plant, which are updated every hour, and the rates database of each of the 750 carriers that GAF uses regularly. Also included is a carrier commitment file that consists of the capacity that each carrier can provide to GAF each day. These files are compiled from faxes or phone calls from each carrier. EDI carrier reporting is coming soon.

Every hour, the model matches orders with carriers and determines the lowest-cost way to satisfy these orders. Dispatchers try to get as close as possible to that plan, but actual availability of carriers does not always match the theoretical model.

"The model is run at the plant level, not at the corporate network level, which would add complexity we are not ready for quite yet," says Schlegel. "The idea was to demystify optimization as a tool. We want to prove that optimization can work. This model is the first step in changing corporate behavior using optimization."

The freight-optimization operation is measured in two ways. First, the model archives what it recommended and what actually happens each day. The system then compares what the model said to do with the actual activity. The "hit rate" on which carrier was actually used vs. which carrier was suggested is now at about 50 percent. The other measurement is how much money was lost by not optimizing every shipment.

The model and the experiment have worked. Freight costs have decreased $500,000 per year. But the benefits go beyond mere cost reductions. "Every day we are finding new ways to save on transportation costs, but just as important as the savings, we are beginning to change corporate thinking and corporate behavior," says Schlegel.

Moment of Truth

Freight optimization, however, is just a small part of the proposed enterprisewide supply chain optimization plan. GAF is still in the process of defining the scope of its plan, but it has identified several elements--including order entry, inventory planning, and warehousing and procurement management--that could be included in the final system.

But folding all those functions into an enterprisewide program will require a significant systems investment ... and an enormous management commitment. At GAF, management is on board with the concept of supply chain management and its benefits. And a supply chain council made up of functional and divisional vice presidents who meet regularly to discuss what can be accomplished and how to proceed already is in place.

That leaves GAF where it is now--at the moment of truth. "We know the costs, the benefits, and the payback. Our supply chain council will make the go/no-go decision based on the potential upside against the cost of implementing the system," says Schlegel.

The decision is not just a matter of spending the money to buy and implement the system, according to Schlegel. Taking the plunge into supply chain optimization would move GAF to a new level of what Schlegel calls "radical strategic change" that could ripple through the company. The payoff of such change should be at least a 15-percent improvement, but these changes could also lead to major alterations in business processes and organization and competitive positioning.

As for what kind of system it will implement, GAF has done preliminary evaluations of all the major supply chain solution providers, and for the forecasting function, Schlegel held a "bake off" among several software venders. These vendors took live GAF data from the company's ERP (enterprise resource planning) system and ran it through their forecasting models. Schlegel says he expects to hold a larger competition for the vendors of supply chain optimization systems soon.

GAF has not decided whether to buy a supply chain solution from one of the major vendors or develop its own. The packaged software is very good, says Schlegel, who has evaluated products from at least five major vendors. But he adds that the vendors are becoming harder to differentiate because the suites that they offer are constantly expanding to take in more and more functions. When evaluating software vendors, Schlegel tries to envision the solution that GAF will need five to eight years from now and then determine which vendor is most likely to get him there.

Schlegel says that GAF's supply chain optimization decision could happen any day, or it could take up to a year. "It's a matter of corporate style and attitude," says Schlegel. "I am more concerned about doing the job right than about making a quick decision. We know what our goals are. We want complete integration of our efforts. We want more enterprisewide thinking and more enterprisewide measurements. We want cost-optimization tools that will lead to strategic change and competitive advantage."

Company: GAF Materials Corp.

Challenge: Reduce transportation costs on 200,000 shipments per year

Solution: Install a freight-optimization decision-support system

Payback: Freight costs have decreased by $500,000 annually