In a global supply chain, there are any number of things that can go wrong. And eventually, they do.

Strikes, storms, congestion, disease, terrorist attacks, supplier failures, missed deliveries, raw-materials shortages: the possibilities seem endless. What's more, as manufacturers move to outsource production to low-cost countries, the list grows even longer. By stretching out distances and adding partners to the mix, companies increase the likelihood of disruptions.

There's no better illustration of Murphy's Law than the day-to-day reality of managing the flow of product from order to delivery. Yet few companies seem to have approached the problem in a comprehensive manner. In a study conducted last year by the Aberdeen Group, 82 percent of respondents said they were concerned about supply chain resiliency, and the implications of failure in their supplier or logistics networks. In fact, the same percentage said they had suffered disruptions that caused financial hardship in the previous 24 months. But only 11 percent were actively managing the risk.

That might finally be changing. Events of the past few years, including the terrorist attacks of 9/11 and severe congestion within the U.S. transportation network, constitute one reason. Add to that the growing interdependence of supply chains, as companies come to rely on outside partners to keep the pipeline flowing. The result, according to Mark Hillman, research director of AMR Research: "The whole topic of risk management is starting to become more of a focus for supply chain executives."

Their priorities, however, change from year to year. In May 2006, an AMR survey identified supplier failure as the biggest potential threat to business. Twenty-eight percent of respondents cited that issue, followed by 17 percent for strategic risk and 15 percent for natural disasters. One year later, supplier failure had slipped to number four or five, with commodity cost volatility -not even in the top six in 2006-topping the list.

The change, says Hillman, reflects an awareness among supply chain executives of skyrocketing prices for fuel and other key commodities. Higher gasoline costs have an impact on the prices of sugar and corn, both of which can be used in alternative fuels, in turn raising the price of beef and other products. Hillman cited one report which found that close to 80 percent of companies are hedging their raw-materials exposure to reduce the volatility of input cost. Others are adopting a similar strategy to ensure continued availability of raw materials. Apple Computer bought up huge supplies of flash memory to support its latest generation of iPods, for example. (Ironically, the move intensified risk for Apple's rivals, who found themselves without enough flash memory to launch competing products.)

In general, Hillman says, executives are shifting their risk focus to matters within their control, such as product quality and the ability to respond to spikes in demand. Natural disasters, geopolitical events and even supplier failures are still vital considerations, but they're viewed as unavoidable factors to which companies must quickly react, not try to prevent.

The theme that runs through many companies' risk management programs is visibility. Managers are working to get a better view of both supply and demand throughout their supply chains, in order to reduce the cascading errors that end up having a devastating impact on customer service - the so-called bullwhip effect.

At the same time, says Hillman, companies must rely to some extent on demand forecasts, imperfect though they may be. He cites the example of Intel, which created an internal "forecast market" by allowing "players" from various functions to issue forecasts in the manner that one might bid for stocks. The goal was to develop a more accurate forecast based on "the wisdom of crowds" - the notion that a collection of experts will be more accurate than any single member of the group.

From the Beginning

Some companies are honing in on the very beginning of their supply chains. Dow Chemical Co. has developed a detailed methodology for assigning risk to each of its raw materials. The program is called PRAM, for Purchasing Risk and Mitigation. It is based on the belief that procurement issues are paramount in heading off glitches in the supply chain. Paulo Moretti, who heads up Dow's global Purchasing Expertise Center, cites an AMR study which found that supplier problems were responsible for a whopping 49 percent of plant shutdowns. (Logistics failures were a distant second, at 17 percent.)

PRAM scores are assigned in two stages, initial and final. For the first, Dow's supply managers or buyers rate each raw material on a 1-to-10 scale, with 10 connoting the highest risk. The score is based on the chief "drivers" behind supply, such as the number of suppliers and plants making each item. The same individuals are then asked to think about how they might decrease the risk, along with the cost of doing so. With that element factored in, the scores for individual drivers-usually numbering between 12 and 16-are added up for a final PRAM, which can be between 1 and 100.

For example, one risk driver might have an initial PRAM score of 10, because a raw material has a single supplier. The Dow manager then explores the possibility of qualifying a new supplier in a low-cost country-say, at a cost of $100,000. As a result, the PRAM for that material drops to 4, with implications far exceeding the expense involved.

Since launching of the PRAM program, Moretti says, half of all assessments have resulted in risk mitigation actions. One Japanese supplier, located in a region plagued by hurricanes, carried $125m of margin at risk; Dow developed an additional source for the product. Another supplier had $80m of risk exposure because the material in question was coming under increasing regulatory scrutiny. Dow came up with a substitute material, reducing both supply risk and procurement cost in the process.

The scores can be tallied at various levels of the organization, whether by buyer, business or region. Moretti has also developed a PRAM methodology for logistics, with separate numbers for barge, marine rail and truck. Next up are PRAMs for maintenance and repair operations (MRO), due at the end of this year, followed by corporate services and information technology purchasing. Ultimately, Moretti views the program as a means, not just of easily assessing risk in Dow's various functions, but of establishing personal goals for employees to reduce risk throughout the supply chain as well.

Dow is acutely aware of its responsibility for the safe movement of hazardous materials. As part of a five-point program, the company is working to reduce risk in its supply chain, by relocating plants, reducing transportation mileage through new sourcing arrangements, or even stopping the sale of certain materials. Other measures include the redesign of railcars handling toxic substances, implementation of training programs for emergency response along selected routes, collaboration with outside organizations to beef up safety, and improvements in track-and-trace systems. By the end of this year, Dow will have GPS sensors installed on its railcar fleet associated with the movement of toxic inhalation materials, according to global supply chain director Dana Mathes.

Restructuring at Motorola

Motorola is another pioneer in the field of risk management. According to Hillman, the company is undertaking a major restructuring of its supply chain, with an emphasis on managing risk and continuity, especially with regard to new product launches. By embracing dynamic scenario planning and capacity modeling, Motorola hopes to reduce supplier lead times and tighten relations with key partners in the chain.

Gerald McNerney, a senior director of Motorola, says risk must be approached from two perspectives: day-to-day business activity, and occasional, large-scale disasters such as hurricanes and plant fires. In the case of the first, it becomes critical to achieve visibility of, and control over, goods moving through the chain. When receiving data, managers must sort out the "noise" from accurate information on changes in supply. During the dotcom crash, McNerney says, some companies didn't respond to real demand signals and ended up with huge amounts of overstock or the wrong mix of product. "It's truly an art," he says, "to be able to make sure you get the right information and make the right decisions."

How well a company reacts to disaster can determine the success or failure of a supply chain, says McNerney. In his 2005 book "The Resilient Enterprise," MIT professor Yossi Sheffi illustrates the now-famous example of Nokia and LM Ericsson, two cellphone makers who reacted in very different ways to a relatively small fire at a Philips semiconductor factory in New Mexico in 2000. Nokia responded to the temporary cutoff of vital components by demanding that Philips immediately supply chips from alternative sources. Ericsson, by contrast, took several weeks to act, causing it to fall behind on a new-product introduction and lose market share, which it never recovered.

"Nobody owns all aspects of the supply chain," comments McNerney. "You're working with partners all over the globe. It requires a real tight relationship, and an acute awareness of how things are changing."

Infor, a supply chain software provider based in Alpharetta, Ga., encourages companies to draw up a profile that highlights the risks of greatest concern. For many, that centers on the increasingly global nature of supply chains, says Andrew Kinder, Infor's director of industry and product marketing for supply chain management. Longer supply lines mean a greater potential for delays and capacity "pinch points." To minimize that exposure, Kinder says, companies should explore a range of alternative supply sources and transportation routes. They should also consider changes in manufacturing strategies in order to take advantage of shifting energy costs, he says.

Communication becomes more difficult as additional parties get involved in sourcing and procurement, says Andrew McGlasson, Infor's director of industry and product marketing. One problem that often arises is the question of accountability for decisions, and liability for the actual goods at any given stage of the supply chain. Buffer inventories can help to offset the risk, he says, but companies must avoid leaning on that solution too heavily, as it weighs down balance sheets. Improved links among key partners, along with well-thought-out contingency plans, can reduce the need for expensive safety stocks.

It's All About Risk

Risk isn't a side issue of supply chain management; it's the entire issue, says John Cummings, chief marketing officer with i2 Technologies in Dallas. "You can really boil the whole equation down to risk," he says.

It's vital, therefore, for companies to engage in detailed scenario planning. They should be able to model the various entities within a supply chain, examining the potential impact of each event, then devise plans for those that impact customer service the most. A key step, says Cummings, is getting closer to demand signals - a strategy that can be tough to implement, given the complexity of today's global supply chains.

One i2 client, ON Semiconductor, implemented a process which allows it to evaluate multiple scenarios on a weekly basis, in order to plan manufacturing capacity. Through the use of automated, rules-based evaluations, the company can react faster to supply chain disruptions. The software includes analytics capability, which supports decisions and helps to identify developing trends.

Getting the big picture is essential, says Kurt Cavano, chief executive officer of TradeCard Inc. Based in New York City, his company makes software that manages procurement. "You need to think about your supply chain holistically," Cavano says. "You should look at vendors as partners in the whole process."

Again, the key lies in achieving visibility across multiple partners - "giving them complete transparency into what's happening," in Cavano's words. Companies can go a long way toward unifying their supply chain partners by eliminating paper communications and putting everyone on a common software platform, he adds.

When it comes to sourcing, Cavano urges a "portfolio approach"-one that defines a set of trusted vendors without relying on a single provider or region of the world. Some sources might be placed closer to the customer, which is a more expensive option on paper but could end up saving money in the event of a supply chain disruption. Creative distribution strategies might also be called for. Cavano cites the example of Burton Snowboards, which receives some imported product at Vancouver, B.C., then ships it by train to Montreal and over the U.S.-Canada border to its distribution center in Vermont. The company can't afford to be without sufficient inventory when the snow season hits, he says.

Visibility of suppliers and carriers is essential, but it doesn't solve the problem of risk entirely, says Paul Hoy, manufacturing industry director with Ottawa, Ont.-based Cognos Inc. Companies don't always have enough advance notification to cope with the unexpected, he says. So they need to equip themselves with management dashboards and scorecards that can identify trends in advance.

Dashboards allow managers to view the progress of their supply chains according to a collection of key performance indicators. In this manner, they get early warnings if suppliers or carriers are underperforming. The system includes a series of event-based alerts that let companies know when their networks are operating outside pre-established tolerances. In addition, Hoy recommends that companies engage in proactive modeling exercises as part of a concerted sales and operations planning process. "That gives customers a really strong what-if capability," he says. "They can do cost-versus-delivery trade-offs."

How Outsourcing Helps

Outsourcing is a major source of supply chain risk, but it can also pose a solution to the problem. Robin B. Shahani heads up supply chain services for EquaTerra, a sourcing adviser based in Houston and London. He says that full-service business process outsourcing, offered by companies such as IBM, Accenture and Capgemini, can help companies to choose the right suppliers in a global supply chain. The use of outsiders for this purpose might appear to be riskier, Shahani says, "but it's not always the case that building or expanding [such capability] internally is a low-risk proposition."

In relying on an experienced independent partner, he says, companies could find themselves better able to lessen the impact of disruptive events such as plant shutdowns. Often those providers have individuals on the ground in sourcing countries, close to manufacturers. "They are doing something you could never afford to do in-house," Shahani says.

Sridhar Tayur, chief executive officer and founder of Pittsburgh, Pa.-based Smart-Ops Corp., offers yet another angle on risk management that companies might have missed. To operational and catastrophic risk, he adds a third category: brand or corporate reputation risk. This often-overlooked element is critical when products are found to be faulty or contaminated. A recent example is the discovery of lead in the paint used to decorate models of toy trains and accessories sold under the name of Thomas & Friends. Toymaker RC2 was forced to recall 1.5 million units when the problem was uncovered. There was also the recent case of a poisonous chemical found in dental products made in China.

At such times, says Tayur, a company's inability to handle the crisis quickly and satisfactorily can damage its brand identity for years, if not forever. The problem becomes more acute as companies turn over production to low-cost countries, where regulations and monitoring standards might not be as stringent as they are in developed nations.

Finally, there is the question of risk at the very end of the trail: the service supply chain. Aftermarket services constitute a multibillion-dollar business, says Tim Andreae, senior vice president of global marketing with MCA Solutions in Philadelphia. For many companies, good service is the key to customer retention. They can't afford glitches, especially when contracts promise replacement parts or repairs within a matter of hours. Add to that the need to support some products for years after they are no longer being sold, and the challenge becomes even more daunting.

The decision on how much inventory to position in the field is essentially a risk management problem, Andreae says. Even the rarest failure can do serious harm to a customer relationship. So a certain amount of safety stock, located relatively close to the customer, is necessary. In addition, companies need systems that account for production lead time, level of service promised and material constraints. In a worst-case scenario, they might be forced to allocate inadequate levels of stock among preferred customers.

Overall forecasting accuracy is not a valid determinant of success, Andreae says. A compound number won't tell the story of whether a company is meeting customer expectations in the event of a supply chain disruption. In a comment that could apply to any link in the chain, he says: "What's important to measure is what service level you actually achieve."