Rise of the Machines – Predictive Analytics in Supply Chain Planning
Editor’s Note: Chainalytics is a leading global supply chain consulting firm and long time ToolsGroup partner.
“Open the pod bay doors HAL…. I’m afraid I can’t do that Dave.” The movie 2001: A Space Odysseywas made in 1968, so the idea of independent computer intelligence is hardly new. But it seems like an explosion in unsupervised computing is now imminent. As long as Skynet isn’t about to become self-aware and launch full-scale warfare on humanity, I’m pretty excited for what the future holds.
Of course, even without autonomous robots walking the earth, the application of predictive analytics to web, social and sharing economy data sets is already revealing a dark side. Groupon keeps offering me deals on precisely the items I had been privately coveting. Facebook is tracking your location in real time unless you opt out. And Uber can figure out who is having an affair. Will the police start arresting people before they commit the crime? Hello Minority Report.
However, I do believe there is incredible upside coming from the use of these techniques in supply chain, and perhaps in no single area more than in integrated demand and supply planning.
My forecast is for better forecasts
Predictive analytics is going mainstream in supply chain planning today. Not tomorrow.
The Chainalytics Freight Market Intelligence Consortium (FMIC) and Demand Planning Intelligence Consortium (DPIC) are two examples of where data science has been conveniently packaged up for consumption by supply chain planners and business owners.
Both FMIC and DPIC leverage quantitative predictive models based on transactional data. These models determine key drivers of the business metrics of interest, and in doing so each service helps identify not only how well the user is doing at managing freight spend (FMIC) and forecasting demand (DPIC), but how well they should be and could be doing in those areas. This intelligence is then used to give the subscribers direct guidance on how and where to drive improvements and what is achievable for their specific business model.
Machine learning, genetic algorithms, and other types of artificial intelligence (AI) are also now in supply chain planning software packages, yielding powerful results. Supply chain software vendors have either founded their platforms on one or another version of pattern recognition technology, or have integrated varying amounts of AI into their products over the past few years.
In fact, in a recent Nucleus Research report on MEIO vendors, all three of the top multi-echelon inventory optimization (MEIO) platforms listed employ a predictive data science technique, while almost none of the other vendors do the same, suggesting that predictive analytics is a major differentiator. Better sensing and better forecasting yields better inventory.
So does the Matrix have us?
All that said, even Hollywood understands that human decision-making wins in the end. Dave ultimately unplugs HAL, the Terminators are eventually defeated and Neo ends the war after only two bad sequels. As predictive analytics becomes commonplace in supply chain planning, human decision making will be enhanced rather than replaced by a new source of intelligence.
About the author
Ben YoKell is Principal for the Integrated Demand & Supply Planning Practice and leads the Demand Planning Intelligence Consortium (DPIC) at Chainalytics. He is passionate about the use of quantitative techniques in supply chain management, and thrives on helping companies leverage optimization and analytics in planning environments with an emphasis on actionable results. In conjunction with the application of analytics, Ben brings to bear significant knowledge of supply chain planning practices and processes for a well-rounded approach to improvement initiatives. Ben can be contacted at email@example.com.