Supply Chain Network Design: Embrace the culture of modeling

We recently helped a consumer products manufacturer take a closer look at its supply network for a family of products being manufactured in China and shipped via ocean container for distribution in the U.S.

The cost savings of “re-shoring,” or shifting production to North America, was not apparent until a number of complex and interconnected factors were considered, namely increasing ocean freight rates, the shrinking differential of labor and other manufacturing costs, as well as the true cost of in-transit and safety stock inventory.

Today, there are a number of “new normals” that are causing transportation and logistics managers to rethink their supply chain networks. The top five are listed below. 

1. Re-shoring: Macroeconomic shifts like the ones described above coupled with increased pressure to position production and inventory closer to points of market consumption, are making re-shoring and near-shoring a more viable supply chain configuration.

2. Global transportation infrastructure: Transportation infrastructure investments like the forthcoming expansion of the Panama Canal are changing the cost and speed of global freight movements and introducing new logistics options.

3. Carrier capacity and freight rate trends: Transportation regulations aimed at safety, like the new hours-of-service (HOS) rules issued by the U.S. Federal Motor Carrier Safety Administration, are affecting carrier operations. 

4. Energy: Continued fuel cost fluctuations are affecting modal choice and spurring investment in technology innovations. Alternative fuels like natural gas are becoming more widely available and redefining the operational landscape.

5. Omni-channel fulfillment and the “Amazon Effect:” Customer-focused fulfillment initiatives that are designed to make products available to buy or return anywhere in the distribution or retail networks are introducing new drop-ship, direct-ship, replenishment, and reverse logistics paradigms.

A fresh perspective
An attempt to internalize these realities within a specific industry or business environment quickly reveals what we intuitively know: We are not very good at predicting seemingly straightforward changes and their effects on our supply chains.

Consider, for instance, the new HOS rules. It was widely touted by the carrier market that reducing the number of on-duty hours would tighten capacity and increase transportation costs by as much as 17 percent. According to an analysis conducted by Chainalytics’ Freight Market Intelligence Consortium (FMIC), rates have not increased much if at all since the rule went into effect earlier this year, and have even decreased in some markets. 

The study examined actual freight costs of more than 100 shippers who altogether purchase $18 billion in freight each year. Any incremental change can largely be attributed to inflation and fuel surcharges.

And let’s not forget the fool’s game that is forecasting fuel prices. The chart below that tracks the forecast for the price of highway diesel fuel against the actual shows the best efforts of the U.S. Department of Energy over the last five years. Simply put, the system is too complex to get it right.

To make matters worse, changes don’t happen in isolation—they occur in conjunction with other complex changes. For this reason, supply chain and logistics managers should focus not on predicting the future, but instead preparing their network for a range of possible outcomes.

Think about the forthcoming Panama Canal expansion and its effect on global transportation movements. How much the new Asia to Eastern U.S. port rates and canal transit fees will be—which are still unknown and subject to market fluctuations—are not the only questions worth considering. When will the various port-deepening projects be ready? How will it affect the price of alternative modes or routes? What will the overall economy look like?

The future as it relates to online and omni-channel commerce holds just as many hard to answer questions. Consumers continue to demand customized products and fulfillment services, and retailers are rapidly experimenting with and deploying alternative supply chain networks to meet this new demand.

We assume that these forces will eventually converge on a small number of dominant designs, but we don’t know what this new era of e-commerce will look like yet. The bottom line: A company must establish an ongoing, scenario-based business process to effectively manage its network. 

How to get started
Supply chain and logistics managers should be constantly reviewing their network infrastructure and transportation flows and testing them in a modeling environment. The goal is to develop an agile network that is robust, provides low cost and high service under the widest possible range of business conditions and can be easily transformed once a key tipping point is reached. 

We recommend the following steps to get started:
1. Set yourself up for success. Consider using an independent partner, one that is deeply familiar with supply chain design, your industry challenges, and the various tools available in today’s technology marketplace. Strategic and tactical planning is too important to skimp on.

2. Build an ongoing, stable competency. Whether it’s in-house or outsourced, the team needs to have enough critical mass to support organizational change management, training, and mentoring. Generally, internal groups are sustainable with five or more people. Groups with fewer than that have a much higher risk of being set back with the promotion or departure of key team members.

3. Use high quality design data.This type of data describes new options in the network. It is data that did not exist historically, such as the costs and capacities of new warehouses and the freight costs of new lanes. Despite the fact that freight accounts for nearly 80 percent of the costs in a supply chain network analysis, many companies still lack the ability to accurately estimate these costs on a large number of lanes. Freight rate benchmarks, like those provided by the FMIC, are increasingly being used as reliable estimates for this critical input.

4. Generate executive support and buy-in. Build confidence in the process, data, and models so executives and managers alike will come to trust and rely on the dynamic tool to support strategic and tactical decision making. The best modeling and analytics money can buy are of little value if the management team lacks confidence in the recommendations.

5. Keep the process and supporting models current with frequent refreshes. The rule of inertia that says “objects at rest tend to stay at rest, and those in motion tend to stay in motion” applies to supply chain and transportation network modeling. Also, keep in mind that jump-starting a discarded model and process is much more difficult and costly than maintaining one. An ongoing, refreshed model can be used to answer ad-hoc questions as well as conduct periodic reviews and planned network analyses. It can also provide differentiated support to sales and operations planning (S&OP) and other cross-functional planning teams.

Case Study: From silos to a culture of modeling
In the absence of a standard, repeatable process for supply chain design, decisions are made in decentralized silos, often with sub-optimal results. One manufacturer of consumer goods realized that it lacked cross-silo decision making. Its procurement, manufacturing, and distribution decisions were being made independently, each with a different set of constraints and data assumptions.

As an example of this separation, one plant in the company’s network had negotiated a tiered pricing arrangement for its inbound resin products. While the top-most tiers provided substantial discounts, the associated quantities were far above the individual plant’s historic volumes. We worked with the manufacturer in building a repeatable process and culture of modeling that incorporated all the relevant stakeholders from procurement, manufacturing, and distribution. 

The network model demonstrated that it was beneficial to leverage the top-tier of resin pricing even if it meant substantially increasing the plant’s production and service territory and scaling back the other plants in the network. This originally seemed counter-intuitive to the planners within each silo, but it made sense when the entire network was considered. The savings from material procurement more than offset the increase in freight costs due to the reconfiguration of the manufacturing network.

Today, the company uses supply chain modeling to frequently analyze network alternatives. It can shift production and distribution on a monthly or quarterly basis as necessary to take advantage of cost and service opportunities. And its in-house modeling team has demonstrated the importance of cross-silo analysis and how a culture of modeling can support and inform strategic and tactical decisions.

Facing the new realities
The realities of the new global economy are a reminder that we operate in a complex and dynamic environment. The future is unpredictable, and the best way a company can manage uncertainty is to continuously assess its supply chain and transportation networks.  By establishing an ongoing business process that evaluates a supply chain’s design and its ability to respond to changing market forces, a company can have confidence in its plans, even under uncertain conditions.

Steve Ellet ([email protected]) is vice president of the supply chain design practice at Chainalytics. Gary Girotti ([email protected]) is vice president of the transportation practice at Chainalytics.