ARTICLES

Managing Unpredictable Supply-Chain, From Superstorms to Factory Fires (Harvard Business Review article summary) 

Traditional methods for managing supply chain risk rely on knowing the likelihood of occurrence and the magnitude of impact for every potential event.

·        For common supply-chain disruptions—poor supplier performance, forecast errors, transportation breakdowns, and so on—those methods work very well.

·        To address this challenge, we developed a model—a mathematical description of the supply chain that can be computerized—that focuses on the impact of potential failures at points along the supply chain (such as the shuttering of a supplier’s factory or a flood at a distribution center), rather than the cause of the disruption.

·        A central feature of our model is time to recovery (TTR): the time it would take for a particular node (such as a supplier facility, a distribution center, or a transportation hub) to be restored to full functionality after a disruption

·        TTR values are determined by examining historical experience and surveying the firm’s buyers or suppliers

·        Our model integrates TTR data with information on multiple tiers of supplier relationships, bill-of-material information, operational and financial measures, in-transit and on-site inventory levels, and demand forecasts for each product.

·        To conduct the analysis, the model removes one node at a time from the supply network for the duration of the TTR., t then determines the supply chain response that would minimize the performance impact of the disruption at that node—for instance, drawing down inventory, shifting production, expediting transportation, or reallocating resources. On the basis of the optimal response, it generates a financial or operational performance impact (PI) for the node, A company can choose different measures of PI: lost units of production, revenue, or profit margin, for instance. The model analyzes all nodes in the network, assigning a PI to each. The node with the largest PI (in lost sales, for instance, or lost units of production) is assigned a risk exposure index (REI) score of 1.0. All other nodes’ REI scores are indexed relative to this value (a node whose disruption would cause the least impact receives a value close to zero). The indexed scores allow the firm to identify at a glance the nodes that should get the most attention from risk managers.

·        The model uses a common mathematical technique—linear optimization—to determine the best response to a node’s being disrupted for the duration of its TTR. The model accounts for existing and alternative sources of supply, transportation, inventory of finished goods, work in progress and raw material, and production dependencies within the supply chain.

Identifies hidden exposures:

The model helps managers identify which nodes in the network create the greatest risk exposure—often highlighting previously hidden or overlooked areas of high risk. It also allows the firm to compare the costs and benefits of various alternatives for mitigating impact.

Avoids the need for predictions about rare events.

The model determines the optimal response to any disruption that might occur within the supply network

Reveals supply chain dependencies and bottlenecks.

Companies can also use the analyses to make inventory and sourcing decisions that increase the robustness of the network.

Contracts with backup suppliers can be negotiated to give a company priority over others should a disruption with the primary supplier occur, which would decrease time to recovery and financial impact.

Prescriptive Actions

Our model provides organizations with a quantitative metric for segmenting suppliers by risk level. Using data generated by the model, we can categorize suppliers along two dimensions: the total amount of money that the company spends at each supplier site in a given year, and the performance impact on the firm associated with a disruption of each supplier node.

Obvious high risk.

They represent 20% of the suppliers but account for about 80% of a firm’s total procurement expenditures, Because strategic components typically come from a single supplier, appropriate risk-mitigation strategies include strategic partnering with the suppliers to analyze and reduce their risk exposure, providing incentives to some suppliers to have multiple manufacturing sites in different regions, tracking suppliers’ performance, and developing and implementing business continuity plans..

Low risk.

Suppliers with low total spend and low financial impact do not require intense risk-management investment. In our experience, most companies effectively manage the minimal risks from disruptions of these supplier sites by investing in excess inventory or negotiating long-term contracts

Hidden risk:

For example, in the automotive industry, a carmaker’s total spend on suppliers of O-rings or valves is typically quite low, but if the supply is disrupted, the carmaker will have to shut down the production line.  Thus, it is critical to ensure that an adequate supply is available. That can often be accomplished using the strategies that apply to the other segments: investing in excess inventory, requiring suppliers to operate multiple production sites, or implementing dual-sourcing strategies.

Companies can use flexibility to deal with hidden supply risks. For example, system flexibility (the ability to quickly change the production mix of plants) allowed Pepsi Bottling Group to rapidly respond to a supply disruption caused by a fire at a chemical plant near one of its suppliers. Similarly, product-design flexibility (in this case, the use of standardized components) enabled Nokia to recover quickly from a disruption of its supply of radio frequency chips caused by a fire at a supplier’s factory. Finally, process flexibility (achieved in this case by adjusting workforce skills and processes) allowed Toyota to quickly restore the supply of brake-fluid-proportioning valves (P-valves) after a major disruption.

Case Study: Ford Motor Company:

Approximately 61% of the supplier sites would have no impact on Ford’s profits if they were disrupted. By contrast, about 2% of the supplier sites would, if disrupted, have a significant impact on Ford’s profits. The supplier sites whose disruption would cause the greatest damage.

Using the model, Ford was able to identify the supplier sites that required no special risk-management attention (those with short TTR and low financial impact) and those that warranted more-thorough disruption-mitigation plans. The results from the analysis allowed Ford to evaluate alternative steps it might take to defuse high-impact risks and to better prioritize its risk mitigation strategies. 

In March 2012, the auto industry was rocked by a shortage of a specialty resin called nylon 12, used in the manufacture of fuel tanks, brake components, and seat fabrics. The key supplier, Evonik, had experienced a devastating explosion in its plant in Marl, Germany. It took Evonik six months to restart production, during which time the downstream production facilities of Ford and other major automakers were severely disrupted. Had Ford managers used our framework prior to this disruption, they would have detected the risk exposure and associated production bottleneck and proactively worked with Evonik to fast-track its plans to bring online a new plant in Singapore, currently slated to begin production in 2015.