AI to Supply Chain Management

By Warwick Hopcroft of Seidor Bluekey


Supply Chain Management can best be understood as a network rather than just a linear process. It is the collection of activities related to the design, production, delivery and service of a product. This complex interwoven and inter-related grid of activities forms the perfect ecosystem within which to harness the power of Artificial Intelligence.


AI + Supply Chain – The future is now


AI can be defined as the use of computers to simulate human intelligence, specifically including learning (the acquisition and classification of information, and reasoning) and finding insights into the data. At the core of artificial intelligence, therefore, is the ability to recognise patterns. This is really significant in the case of big data, where the so-called 3Vs (volume, velocity and variety) can be leveraged to find correlations within seemingly diverse data. For example. Transportation planning is a key component within Supply Chain Management. However, historically only very limited access to data to plan routes effectively was available. By contrast, nowadays there are AI enabled systems that use cognitive technology to track and predict supply chain disruptions based on gathering and correlating external data from disparate sources such as social media, news feeds, weather forecasts and historical data.


Online shopping provides another great example of new game-changing technology. Innovation from firm Sentient uses machine learning to deliver purchasing recommendations to e-commerce shoppers based on image recognition. Rather than only using text searches and attributes like colour or brand, the software finds visual correlations with the items that the shopper is currently browsing, through the use of visual pattern matching.  


Its all about interpreting data


The key ingredient that AI brings is its ability to filter through, and interpret, large volumes of Big Data. Without this ability, much of the information now regarded as so valuable would be completely useless.  


To better understand the value and impact of Big Data, let’s consider a project Amazon is currently busy with. The company reportedly already employs 1,000 people in artificial intelligence, mostly working on ‘Echo’, a wireless speaker that listens to you and speaks back. As Kevin O’Marah, the chief content officer at SCM World, explains, “What Amazon is positioning itself to do is far more ambitious and involves what AI experts call ‘contextual awareness’. This means knowing not only the what, but also the when, why, where and how of consumer need. The long game is all about selling us not just what we want, but what we need, and probably before we realise we need it.” Amazon is developing a complete picture of each customer, and the personal data collected will help future AI applications to know the difference between what you want, and what you need.


Just imagine how much data Amazon needs to have collected, rejected, analysed and interpreted, in order to be able to tell us what we need. It really is true that whoever can extract the most value from their data and information are going to be the winners in the long-run. For this reason, powerful databases like SAP HANA coupled with capable data mining and the analytical platform is the only way to establish a business advantage when handling big data.


Are the lights going out?


Up until now, supply chain innovations have focused largely on repetitive, transactional processes both in warehouses and production lines. This is, however, starting to change. We’ve entered an era where it’s no longer just a case of swapping out humans for machines. Robotics process automation (RPA), is an emerging form of clerical process automation technology based on the notion of software robots or artificial intelligence (AI) workers. It goes beyond physical systems and provides the glue that when looked at from a supply chain perspective integrates multiple systems dedicated to order-taking and fulfilment. The potential impact of this on work as we know it is almost incomprehensible. Price Waterhouse Coopers estimates that businesses could automate up to 45% of current work, saving $2 trillion in annual wages, taking us ever closer to a world with the so-called ‘lights out’ factories and warehouses.


Finding simplicity in complexity


Clearly, the promise that AI holds, and the impact it is likely to have on businesses are both exciting and far-reaching, but it is not just smooth sailing. Moving into an AI led system within a supply chain is complicated and has often borne less than sterling results. Typically, and AI system requires an extensive team of planners for its implementation which is likely to comprise of complex engines running at each step of the process and at each node in the supply network.


At a human level, there are personalities and the resulting conflict between functions and partners which needs to be navigated through and potentially huge opportunities that go unnoticed because they are embedded within the network itself.


But make no mistake, AI will create value in unexpected and unprecedented new ways. The challenge lies in anticipating these changes and then ensuring you are able to capitalise on them, up and down the supply chain.