Speed is a common measure for many of today's supply chains. We all want to know how quickly we can respond to the customer's request. Speed of response is one of the main drivers behind the current thinking on the Demand Driven Supply Network (DDSN) and many other recent supply chain innovations. One important thing about "speed" is that it comes in at least two flavors, "fast" and "slow". Many of today's best supply chains use both. A key supply chain opportunity is to know when to use which speed for transportation to meet customer service targets at the lowest total cost.

You might ask yourself, "Why, if my supply chain is supposed to be quick to respond to customer requests, would I want to include any speed other than "fast?" One "fast" answer is cost. We see this cost almost every day in our personal lives as well as our professional lives. When we purchase something off the internet, we always have the choice of "standard shipping" for a certain price (sometimes free) for delivery in a few days, or "premium shipping" that can have the package on our doorstep the next morning. However, this next-day service almost always costs us more. Likewise, when a supply chain professional chooses rail for his/her long-distance shipments the costs are substantially lower than they would have been for truck, but at the cost of a longer transit time.

Because of this cost differential, many of today's supply chains use multiple modes (fast and slow) of transportation. The slower mode of transportation is usually applied to lower cost materials that have a longer shelf life and are consumed in a predictable pattern. That is, their demand can be reasonably accurately forecasted. However, in practice there are often significant numbers of shipments of this same material that are made using faster, more expensive modes of transportation. Why is this, and how can we minimize it in our supply chains?

A major factor contributing to the freight premium seen in these types of lanes is the fact that most supply chains are planned with "steady state" in mind, using average demands, average transit times and average supply capacities. Steady state looks great on paper, but rarely happens for prolonged periods in the real world. Thus, we find that natural variation in customer demand, transit time or manufacturing capacity can create low inventory situations that require expedited shipments to avoid a stock-out. In addition, unplanned surges in demand, transportation interruptions (like port congestion, strikes or storms) and temporary shortfalls of supply perturb the system and push us to expedite shipments that are supposed to move by a cheaper mode of transportation.

Reducing the premium freight caused by the natural variation in demand, transit time and supply should happen at the supply chain design stage. Using the right blend of statistics, modeling and experience will result in a much more robust supply chain that balances the additional cost of inventory, and logistics assets against the high cost of premium freight. Dealing with the issue of large, unanticipated perturbations to the supply, demand and transit times is another kettle of fish.

This issue may best be explored by using an example. Profit Point was retained by a large specialty materials manufacturer to help solve this precise problem. The specialty materials company manufactured a number of key raw materials at a large, integrated site on the US Gulf Coast. From there, they were shipped to nine company locations around the US and Canada as well as to a number of external customers. Although there were multiple products manufactured at the site, the level of integration among the products was such that an upset in one part of the process could impact the supply of a number of these raw materials. The preferred mode of transportation for all of the material was by rail in tank cars. However, each year the company was spending millions of dollars in premium freight to move the materials by tank truck to the very same sites. The root causes for the premium freight were:

1. The manufacturing site was required to operate very close to its instantaneous maximum capacity to meet demand. Thus, any small interruptions in supply had large ripple effects through the system because "catch-up capacity" was almost non-existent to rebuild inventory. (This is why you need significantly more inventory when you operate so close to capacity, but that's the subject of a future article. )

2. Rail transit times were extremely variable, particularly to sites in the Far West.

Because neither of these problems would be solved in the foreseeable future, we needed to develop a customized solution that would minimize their premium freight costs while continuing to deal with the ongoing perturbations in the supply chain. In other words, how could we manage the transition from all rail shipments to partial or full truck shipments and back to rail at the minimum cost while meeting customer service requirements?

Using a customized heuristic algorithm, Profit Point developed a finite capacity scheduling application that created a product/customer-site specific schedule of tank cars and tank trucks that maintained minimum safety stocks at the nine company site and met customer service goals for the external customers while minimizing the total freight spend. With this new tool, the manufacturer cut the premium freight cost dramatically and improved overall customer service.

With the tool, the scheduler could create much better schedules in a fraction of the time required with the spreadsheet approach she had been using. (Those of us who have worked on the plant floor remember that "optimum" can be loosely defined as "the first schedule that works," when you're up to your eyeballs in alligators with people calling to find out when they're going to receive their next shipment. The new scheduling algorithm was able to look at hundreds of schedules that would work and choose the best one.) She was able to quickly orchestrate the moves of certain lanes into trucks and then back into rail cars as conditioned changed either on the supply side or on the transit-time/demand side of the process.

If you'd like to find out more about managing multiple transportation modes in common lanes in your supply chain, please call us at (866) 347-1130, or send us an e-mail using the following link: http://www.profitpt.com/contact.asp.

Labels: Profit Vehicle Router, Routing, Transportation