AI in the simplest terms is the creation of “general intelligence” found within humans, fundamentally imagination and memory, without the constraints of traditional thinking or emotions. In the latest report written by the Capgemini Research Institute, the use of AI is explored in manufacturing operations. Many organisations have embarked on the journey towards using AI to drive cost reduction by enhancing quality and improving productivity, but is it delivering full potential?
According to our report, Europe is leading in the implementation of AI in manufacturing with over half of the organisations interviewed for this report having implemented AI solutions. However, AI is not being exploited to its full potential yet. Although there is the ability to apply AI across the breadth of manufacturing operations, initially the same few use cases are being used across industries – quality inspection, predictive maintenance and demand forecast and planning. It is easy to see the logic of how AI lends itself to these areas, but the big question is what next? Can the benefits of AI be delivered at scale beyond these areas? Or has AI found its niche in manufacturing?
In order to get the best out of AI, there are three key consideration points. Firstly, identify a killer application to fit with your industry, clients’ needs and strategy. Secondly, create a clear business case for the solution. A lot of solutions fail before rollout as they fail to prove the value. Finally, data, this is one of the most important components. A robust data strategy is required to support the business case. The challenge sometimes is that there’s not enough data available for machine learning. However, it’s not just tackling the technical challenge, it’s how to overcome the people and cultural change required.
The report shares how Airbus is using AI to realise benefits across its manufacturing operations, but also highlights how some of the key challenges for scaling AI are linked to trust. The big question is how to support humans to trust the insights provided by AI when the results cannot always be explained due to the “black box” nature of AI. A recent Nigel Thomas’ blog on Smart Factories and Industry 4.0 explored the topic of “who is designing the role of the digital me?”. Embarking on an AI journey means you have to allow for “self-learning” time. As AI “self teaches” itself and learns, the accuracy and levels of insight mature. As humans we are good at making decisions. We (humans) may not always feel there is time to wait for AI to evolve, leading to the desire to return back to previous approaches.
The adoption of AI at scale is still rare. Whilst the business case for AI is clear – improved quality, cost reduction and the associated sustainability impact of reducing resource consumption, there are several critical success factors required to take AI beyond the pilots and proof of concept stage. Of all of these, one of the biggest issues to overcome is the role of the human. Until organisations tackle this, it will remain one of the key limiting factors.
Claire Wallin is leading Aerospace and Defence projects in Capgemini Invent. Claire has extensive experience in Defence supply chain and transformation programmes. Her passion lies in operational management and transformation where she can apply her restructuring experience and the expertise gained whilst working in the automotive sector.