Data analytics is the science of examining raw data to help draw conclusions about information. When applied to the supply chain it is often described as supply chain analytics.
One category of data analytics is known as predictive analytics, which uses data to predict trends and patterns; it’s often associated with statistics. In the supply chain, predictive analytics can be used to forecast future demand or to forecast the price of a product.
An interesting application of predictive analytics is being taught at the Fashion Institute of Technology (FIT) in New York City in the use of weather forecasting in the fashion industry (“The Next Fashion Trend: Weather Forecasting”).
While a few companies such as Zara’s (which designs, manufactures and sells fashion apparel) have been very successful in reducing lead times for the apparel they manufacture and sell through their retail stores, the fashion industry in general has long lead times with merchandise often ordered months in advance based on what the weather typically is at that time of year.
Increasingly erratic weather fluctuations due to global warming (last winter was the warmest on record) have been putting fashion designers and clothing retailers on the defensive in terms of forecasting demand. In fact, some retailers have hired climatologists to help them with predictions.
Some designers have been proactive in this regard by offering a range of fabric weights in their collections (Michael Kors) or promoting seasonless clothing with pieces that can be layered on or taken off (Vince).
FIT’s predictive weathering class uses various causal relationship models such as regression analysis to integrate weather into fashion forecasts—something that perhaps wasn’t considered in the past.
Using predictive analytics to consider the impact of weather and other external factors can help fashion and retail companies statistically assess how external factors like weather impacts their business and can result in improved forecasts, minimizing disruption and maximizing profits throughout the entire supply chain. This is something that many industries may want to consider when it comes to forecasting demand.