The speed of business is now computer speed and supply chains need to run at the same speed as the rest of the business. Fortunately, cognitive technologies are maturing just in time to achieve that aim. Cognitive computing is a sub-set of artificial intelligence (AI) combining machine learning, natural language processing, and advanced analytics. Even though there are differences, the terms AI, cognitive computing, and machine learning are terms often used interchangeably. What they have in common is the ability to provide insights drawn from massive amounts of data. For the purposes of this article and for the sake of simplicity, I’m going to lump them together under the cognitive technologies rubric. Supply chain professionals are not as interested in how cognitive technologies work as they are in what cognitive technologies can do for them.

The Benefits of Cognitive Technologies

Business Intelligence and Decision-making

The list of benefits provided by cognitive technologies begins with decision-making. Bain analysts, Michael C. Mankins and Lori Sherer (@lorisherer) observe, “The best way to understand any company’s operations is to view them as a series of decisions.”[1] This is just as true for supply chain operations as it is for any other business operation. “Today,” writes, Nilam Oswal (@nilam_oswal), a software analyst at SoftwareSuggest, “businesses are entering into a new era ruled by data. AI, specifically, is gradually evolving into a key driver that shapes day-to-day business processes and Business Intelligence decision-making. Thanks to advances in cognitive computing and AI, companies can now use sophisticated algorithms to gain insights into consumer behavior, use the real-time insights to identify trends and make informed decisions that give them an edge over their competitors.”[2]

Forecasting and Planning

Machine learning is the most widely used AI tool in business. Steve Banker (@steve_scm), Service Director for Supply Chain Management at ARC Advisory Group, notes machine learning has been used in the supply chain field since at least 2004.[3] Like other AI tools, machine learning requires a lot of data. Banker explains, “For machine learning to work well, it needs to be a big data application.” To state it another way, there is a symbiotic relationship between big data and cognitive technologies. Forecasts have always relied on data; but big data and cognitive technologies can raise the forecasting bar. Banker explains, “In addition to doing forecasting based on historical sales, consumer goods companies leverage other data sets such as their retail customer’s point of sale, recent shipments of products from their warehouses to their stores, the retailer’s orders, syndicated data, and store inventory. Many of these data sets are accessed daily, or even several times a day, so the dynamic nature of demand is captured to a much higher degree than traditional forecasting techniques.”

Because cognitive technologies work at computer speed, Banker notes, “The engine is making many forecasts simultaneously in different planning horizons. … Then every day, all of these forecasts at all locations and all time horizons are done again.” Although that sounds like overkill, Banker explains, “Different forecasts are used in different time horizons by different groups in the company.” Because those different groups are all operating from the same dataset, they remain aligned even though they are using different time horizons for their forecasts. And, Banker reminds us, “The machine learning engine does not rest on its laurels. It continues to monitor the accuracy of all forecasts, continues to reweight which data sets are used, and which algorithms work best, at all locations in all time horizons.” Alexa Cheater (@Alexa_Cheater), Product Marketing Manager at Kinaxis, encourages the use of cognitive technologies. She explains, “Amplify the value of your existing processes and people with machine-assisted planning, which can help you bridge the knowledge gap between experienced and inexperienced planners, and gain real-time recommendations based on historical and current data analysis.”[4]

Advanced Analytics

The common thread in areas discussed above is advanced analytics. Oswal notes there are basically three types of analytics being leveraged by companies:

• Descriptive analytics: “It does precisely what the name implies: description. It summarizes raw data and breaks it down into something that can be interpreted by humans. Descriptive analytics enables companies to understand past behaviors and learn how it can influence future outcomes.”
• Predictive analytics: “This ‘predicts’ the future. Predictive analytics enables companies to have future insights. Although no statistical algorithm can give 100% prediction, organizations are using these analytics to forecast future events. This system relies on ‘best guesses’ since its foundation is based on probabilities.”
• Prescriptive analytics: “A relatively new but robust field that enables users to prescribe various possible actions and advise accordingly towards viable solutions. Prescriptive analytics is all about providing advice. These AI-powered analytics not only predict what will happen but also explain why it will happen.”

Although advanced analytics are of obvious benefit to planning and marketing, Cheater notes they can also be valuable in operations and risk management. She writes, “Improve your supply chain visibility and risk insight by using AI to track and predict possible supply chain disruptions based on inputs and correlations across multiple data sources, including weather forecasts, news and even social media.”

Staying Afloat in an Ocean of Data

The rise of Internet of Things (IoT) means companies are going to find themselves adrift on an ocean of data if they fail to leverage cognitive technologies. As I noted in a previous article, “The distributed sensors of IoT depend on powerful new brains to fully realize their potential. Already, 71% of companies are collecting data from IoT devices, anticipating future initiatives. The challenge is weeding through all that data to find the actionable intelligence that decision makers need. That’s where the embedded advanced analytics and data-crunching power of cognitive-computing platforms can really push intelligence into the supply chain, accomplishing the ‘nearly impossible.’ At the same time, the long-held dictum of ‘garbage in, garbage out’ remains as pertinent as ever. All that data, from all those systems, has to be harmonized and indexed before it can be processed into something useful. A cognitive computing platform functions as the conductor orchestrating all this data into a harmonious whole.”[5] Staff members at Supply Chain Review note, “Every process and transaction automatically generates data. It is difficult, near impossible, for humans to identify useful patterns within the deluge. “[6] They continue:

“Knowledge that human experts derive from data and interpret based on their own experiences can be fallible. Conversely, the Self-Learning Supply Chain’s data-driven knowledge extraction uses algorithms to identify and condense patterns in data into useful information for planning. It then analyzes the data continuously to keep it up-to-date even as conditions change. Business reality and powered by world-class optimization technology, the self-learning supply chain has the ability to intuitively learn and replicate the logic and reasoning of the best decision-makers in the company. It prescribes actions and generates optimal plans that work in the real world.”

What they call “the Self-Learning Supply Chain,” I call the Cognitive Supply Chain.

Summary

Cheater concludes, “Companies are making big strides in developing and deploying real, practical applications of AI in supply chain management and planning.” Oswal agrees and she believes cognitive technologies “will keep them competitive in the tech-powered business landscape.” I predict, once cognitive technologies are in use, supply chain managers will discover more use cases for making the Cognitive Supply Chain even more effective and efficient.

Footnotes
[1] Michael C. Mankins and Lori Sherer, “Creating value through advanced analytics,” Bain Brief, 11 February 2015.
[2] Nilam Oswal, “How AI is Transforming Business Intelligence,” Dataconomy, 19 February 2018.
[3] Steve Banker, “Machine Learning In The Digital Supply Chain Isn’t New,” Forbes, 26 September 2017.
[4] Alexa Cheater, “An executive’s guide to AI in supply chain management,” 21st Century Supply Chain Blog, 27 March 2018.
[5] Stephen DeAngelis, “Cognitive Computing, Blockchain and IoT are revolutionising the digital supply chain,” Supply Chain Digital, 22 March 2018.