While supply chain disruptions are down from the high levels of 2021 and 2022, companies continue to strive to be more adaptable and resilient.
One key barrier: continued poor results from planning efforts, caused by flawed forecasting, which results in delivery delays, inventory levels that are very out of sync with demand, and disappointing financial performance.
Supply Chain Digest Says... |
|
|
What else would you expect. How can inventory and production decisions be made effectively when demand forecasts are widely off?
That’s the view of five authors - Narendra Agrawal of the Santa Clara University’s Leavey School of Business, Morris Cohen of the University of Pennsylvania’s Wharton School and founder of AD3 Analytics, Rohan Deshpande of Cerebras Systems, and Vinayak Deshpande of the University of North Carolina’s Kenan-Flagler Business School – recently writing on the Harvard Business Review On-line.
Of course, poor forecasting is a perennial problem for thousands. But these five authors claim to have largely solved the problem. How?
The five say use of machine learning and historical data can generate superior recommendations for supply chain decisions. They say that while current machine-learning efforts often focus on trying to create more-accurate forecasts, their technique focuses on making actual decisions.
They call this new methodology optimal machine learning, or,OML).
It involves “using artificial intelligence technology to create a mathematical model that takes key data inputs related to the supply chain (the nodes of the network, their locations, sales and shipment transactions, financial parameters, marketing promotions, logistical and capacity constraints, and so on) and links them to planning decisions (what quantities to produce, for example, or what levels of inventory to stock at each location),” the five write.
The approach also can consider variables such customer service targets, budget realities, and supply chain constraints.
OK, sounds good – but not all that different than what supply chain planning software vendors have been promising for decades. What’s different here?
The five say the traditional forecasting approaches use something called predict-then-optimize (PTO), meaning a forecast is created based on historical data and other inputs, then fed into mathematical models of the supply chain network in order to generate final inventory stocking decisions.
This approach, the five say, fails for various reasons. Those include:
1. There is no single forecast that all parties agree to use for decision-making: apparently sales and operations planning still has a way to go.
2. The objectives of the various stakeholder groups in the planning process are not aligned, which leads to biased and suboptimal decisions: the famous Bullwhip Effect is one consequence.
(See More Below)
|
CATEGORY SPONSOR: SOFTEON |
|
|
| |
|
|
3. The methods for deciding how to optimize inventories are flawed: inadequate supply chain models, simplistic algorithms, and more.
Another challenge is well-leveraging the vast amounts of data companies now generate.
 “Unfortunately, for companies with global supply chains, accessing and consolidating that data remains a mega challenge,” the five write.
Ineffective scenario planning is another key barrier, the five say, even though, SCDigest notes, the
technology to do this has improved dramatically in the past 10 years.
Perhaps the issue: scenario planning “often lacks sufficient detail to be useful,” the authors say.
The authors proposed a new paradigm, based on three pillars, as follows:
Decision-support engine: OML replaces the forecast-based, two-step planning process with a one-step process that connects input data directly to supply chain decisions. Let’s just say it involves a rather gigantic data model.
Digital twin: A key requirement for the OML decision-support engine to work is a detailed digital representation of the entire supply-chain network, all material flows, and the decision-making processes of all involved parties.
End-to-end data architecture: The OML decision-support engine and the digital twin require a data storage system that works in conjunction with all existing database-management systems throughout the supply chain (those for the company’s operations and those of suppliers, distributors, and customers).
In conclusion, “OML allows companies to base decisions on historical and current supply-and-demand information rather than just more-accurate forecasts,” the authors say, adding that “It gives them a tool that can help them reduce costs and increase revenues, profits, and customer satisfaction. It enables them to test strategies for mitigating risks, making it easier to choose the best ones.”
The full article, with two case studies, is available here: How Machine Learning Will Transform Supply Chain Management.
Do you have any thoughts on OML? Let us know your thoughts at the Feedback button below (email) or in the Feedback section.
|
