Artificial Intelligence (AI) has been in the news for all the wrong reasons lately, with Microsoft’s attempt to appeal to millennials using a Twitter chatbot proving what many have always feared about AI and completely going off the deep end. Of course, this doesn’t say as much about AI as it does the calibre of millennials on Twitter - and the judgement of Microsoft’s engineers for thinking it was sensible to have them shape an artificial personality - but it is still worrisome.
While many may regard this failure as a warning against giving machines autonomy, however, it’s a warning that’s highly unlikely to go unheeded. AI is already changing our world, with the machine learning algorithms that run it taking control of huge swathes of our daily lives without most of us even realizing. Its rapid growth over the next decade or so will completely change both the nature of employment and the face of business, and strategies need to be in place now to ensure the technology is exploited to its full potential. In a recent survey of Accenture’s clients, 70% of executives said they are significantly increasing investments in AI compared with two years ago. One of the most obvious places to start is the supply chain, where we could ultimately see full automation.
In Gartner’s recent ‘Predicts 2016: Reimagine SCP Capabilities to Survive,’ the research firm revealed that their recent survey had found supply chain organizations expected the level of machine automation in their supply chain processes to double in the next five years. Technology and digitization have already had a major impact on supply chains, and much has been made of the likely impact of the Internet of Things in the coming years. The rise of AI should work alongside this to completely remove the need for humans in the supply chain, which could, if handled correctly, improve safety, efficiency, and enable complete transparency between supply chain partners that will allow inventory to be far more precisely handled, greatly helping to cut costs. It will also give rise to new profit streams from capital equipment, and better environmental performance for existing infrastructure.
Transportation remains the key component of supply chains, and autonomous vehicles are probably set to be the AI technology to make the most profound impact in the field. A self-driving car is any vehicle that does not require the input of the occupant to steer, brake, or accelerate, but rather, uses machine learning algorithms - the basis of AI - to understand its surroundings and take action. The major tech giants, particularly Google, have made it a major priority, and they are already creating and developing such vehicles with an exceptionally low accident rate. The market for self-driving cars and trucks is expected to grow to more than 10m vehicles by 2020, representing an annual growth rate of 134%, and it’s a matter of time before we see them on main roads. This will, in turn, reduce accidents, particularly in areas such as haulage, where the long hours on the road mean drivers run the risk of getting tired and falling asleep at the wheel.
In factories, Siemens is one company that is already using AI and automation extensively. Their lights-out factory has automated some of its production lines to the extent that they can run without supervision for weeks at a time. Theoretically, this could work alongside autonomous vehicles to enable goods to be manufactured and transported to another AI-run factory with no need for human involvement whatsoever.
AI should also mean that traditional models for handling inventory can be used in conjunction with sophisticated algorithms to increase the speed of computation, and it will likely augment this process by generating new features to run such models on. Danone, for example, is already using analytics with machine learning capabilities to analyze their demand planning—gains, which resulted in a rapid 20% reduction in forecast error and a 30% reduction in lost sales.
The idea of a supply chain run without humans raises justifiable fears around rising unemployment, and the dangers of losing control of AI evidenced by Microsoft’s chatbot. However, when used effectively, when algorithms handle order commitments or production planning, human judgement must still be involved in the process. The Siemens’ factory is a good example of why this is likely to be wrong, with more than 1,150 employees supporting their systems, automation has clearly not given rise to mass unemployment. AI will still need careful monitoring, and the knowledge of experienced logistics and operational professionals to ensure that it is being used to its maximum potential.