A company’s ability to evaluate multiple “what-if” scenarios and make rapid decisions with a high level of confidence is what keeps supply chains moving during disruptions like COVID-19. But what do you do if you don’t have an existing network model? Here’s how to get started.
My colleague Jeff Zoroya recently explained how you can repurpose your network model to make rapid decisions. Jeff encouraged companies to refresh their out-of-date models and bring them quickly up to speed to help make quick tactical decisions. However, if you don’t have an existing network model, then you will need to start this process from the beginning—and quick.
An investment in supply chain network modeling yields benefits way beyond its cost. The ability to tactically allocate capacity and cultivate flexibility in sourcing, manufacturing, and distribution requires an end-to-end understanding of the relationships and trade-offs at play. This in-depth knowledge of current conditions can also help you develop more strategic, long-term plans as your business and customers change over time.
Leveraging network modeling tools
Advanced enterprise-level data acquisition, analytics, and optimization tools are the supply chain modeler’s greatest allies. The capability to capture, cleanse, and analyze data has vastly improved in recent years as supply chain management has become a comprehensively digital exercise. If you don’t have this level of power at your disposal, now’s the perfect time to consider making the investment. However, with or without a suite of modeling tools, the real opportunity to build confidence in your network model lies with the people and process.
Building the right team
- Communicate the objective: To ensure everyone is on the same page, develop and state clear goals. For example, do you want to examine a realignment of your distribution network imbalances to deal with demand uncertainty? Or would you like to explore increasing manufacturing flexibility through new locations or sources?
- Define and prioritize your goals: Determine where most of your challenges are, and prioritize your objectives. Be consistent about defining your actions around the objective – it’s essential to explain precisely what needs to be solved.
- Mobilize your team: Assemble a cross-functional team wherein individual team members have a complete understanding of their respective roles. A team with a mix of operations, planning, and data science talent will help you make better decisions faster.
Creating and deploying the supply chain model
As you begin to create your model, remember that the goal of network modeling is to build confidence in the recommended outcome. However, like many other areas of business and life, there are diminishing returns when it comes to increasing the accuracy of your model and investing additional time and effort. Thread this speed vs. accuracy needle carefully as you work through the following steps:
- Compile relevant data: Look for good data, not perfect data (which, in any case, might not ever be available). Transactional data like demand history and sales orders are probably in decent shape, but your customer master data may not be. Deploy resources quickly to address issues in your datasets.
- Monitor dynamic inputs: Especially during a crisis, managers may receive incomplete or conflicting information, so be sure to carefully evaluate input data to assess and account for their possible dynamic nature. And don’t treat these inputs as static, which they seldom are.
- Get a working baseline: Quickly establish a baseline model that captures your existing situation as accurately as possible. Anything within a range of 90-95% accuracy is considered the gold standard.
- Test different scenarios: The recovery scenarios for COVID-19, as polled by the World Economic Forum, find an increasing number of characters used to represent it, including a “V,” “U,” “L,” “W,” and even a “swoosh.” Don’t wait to have a perfect baseline before developing the scenarios and/or hypotheses you want to test. There is much more value in modeling as many scenarios as possible based on a 95%-accurate benchmark than modeling one or two scenarios on one that’s 97% accurate but took much longer to achieve. (And that 97% accuracy is just an estimate anyway!)
- Sensitize, sensitize, sensitize: You can further improve your analysis and outcomes by sensitizing your critical inputs. Let’s say your estimated demand levels don’t materialize. You can compensate for data that remains unknown by defining sensitivities for those inputs and then reviewing the effects on the output. For example, your post-pandemic distribution network may be insensitive to demand/forecast within ±35%. In this scenario, a small change in transportation rates over your expected levels may suggest changes to your future network. Sensitization will show you what actions you need to prioritize.
- Implement your design: At this point, you should have generated the executive confidence and buy-in necessary to choose your go-forward path. To implement your design, start by creating a project roadmap based on delivered value vs. effort. Over time, and depending on the value/effort trade-offs you’re willing to make, you’ll realize results ranging from quick wins and tactical changes to strategic transformations as part of your optimized roadmap.
Amid the COVID-19 pandemic, the words of Nobel Laureate and economist Paul Romer have never seemed more prescient: “A crisis is a terrible thing to waste.” Opportunities emerge, and creativity and innovation are inspired in times like these. There is no better time than now to develop your supply chain modeling and decision support capabilities quickly. Make sure your company is primed for the growth opportunities to come.
Change is an opportunity for growth – if you seize it. Chainalytics’ combination of top supply chain talent, proven methodologies, and exclusive market intelligence consistently puts our clients ahead of the curve. Reach out to us and see how Chainalytics can help refine your supply chain’s reliability and efficiency and position you for a robust resurgence.
Milind Kanetkar is a Sr. Manager in Chainalytics’ Supply Chain Design practice and is based in Bangalore, India. Milind has a rich and diverse mix of experience in supply chain optimization, planning, and operations across multiple industries.
