Demand forecasting has changed drastically since 2020. Traditional models are becoming obsolete due to the unprecedented COVID lockdowns, supply chain issues, the unstable economy, the first major war in Europe since WW2, and an ongoing energy crisis. This makes internal and external data essential for demand forecasting, which optimizes inventory levels and reduces costs.
In light of this, Solventure’s own Gylian Verstraete (Manager Data Science) went on the 10th episode of the Miebach Consulting podcast to offer an alternative to these traditional models. He was joined by Gustavo Escobar, Manager at Miebach, partner of Solventure and a global supply chain consulting company, to discuss forecasting with advanced analytics.
This conversation couldn’t have come at a better time, as supply chain issues and COVID-related pent-up demand have led to large backlogs at companies across industries, making them prone to the phenomenon called the supply chain bullwhip.
Staying competitive in a changing world
Verstraete: “Companies typically use traditional statistical techniques in their sales forecasting process, which means historical trend and seasonal patterns will be extrapolated into the future... However, relying solely on these techniques can be risky, especially in times of supply constraints and disruptions, as your historical demand pattern will no longer be a good representative for the future. That’s why I emphasize that companies need to use external data in their demand planning process to gain more visibility into their supply chain. This can include macroeconomic indicators such as production rates, oil prices, weather information, and even events like the Chinese New Year or labor strikes. Many industry-specific indicators are available too: vehicle sales for automotive, building permits for construction,.. ”
Escobar: “ I would also like to add that companies need to be proactive in their approach to demand planning and forecasting. By monitoring external data, they can detect early warning signals of changes in demand or supply constraints and adjust their plans accordingly. This can give them a competitive advantage over companies that are still using traditional forecasting methods.”
Not everyone seems ready for predictive forecasting tools
Escobar: “Most companies are currently not using predictive or prescriptive tools in their forecasting processes. This highlights the need for companies to invest in the right technology and people to support their demand planning efforts.”
Verstraete: “And even if the right technology and people are in place, companies need to expand their view of demand planning beyond just their own internal data. They need to work with other divisions in their supply chain to get a more complete picture of demand and inventory levels. Furthermore, companies need to be aware of the potential pitfalls of relying solely on internal data. Why only look at internal demand & inventory levels and not at other stakeholders in the value chain? An example that highlights this clearly is one of a retailer of second-hand vehicles in the US that saw a surge in demand during the COVID-19 pandemic. However, as lead times for new vehicles became shorter, demand for used vehicles decreased, leading to lower sales volumes, lower selling prices, and ultimately lower profits. The company also needed to ensure the existing car park (inventory) could still be sold above the purchasing price. By monitoring external data such as vehicle sales, vehicle production, and consumer prices for both new and used vehicles, the retailer could have detected this trend earlier and adjusted their plans accordingly.”
Get a competitive edge thanks to external forecasting data
In conclusion, using external data in demand planning and forecasting is essential for companies that want to stay competitive in today's rapidly changing business environment. Companies need to invest in the right technology and people to support their demand planning efforts and work with other divisions in their supply chain to get a more complete picture of demand and inventory levels.
By doing so, they can detect early warning signals of changes in demand or supply constraints and adjust their plans accordingly, giving them a competitive advantage over companies that are still using traditional forecasting methods.
