More and more manufacturers are using AI to increase their operational efficiency, according to Valtech’s latest study “The Voice of Digital Leaders in Manufacturing” (VODL). However, with AI’s potential extending far beyond short-term efficiency gains, promising a transformative impact on the entire business model for many manufacturers, is the technology being utilised enough?
Currently investment is not where it should be, and manufacturing leaders seem to struggle with knowing where to start. Overall, there appears to be a discrepancy between the industry’s aspirations and reality. According to Valtech’s study, many manufacturers see AI as a trend that will have the greatest impact on their business.
However, AI was at the bottom of their list of digitisation priorities in 2024 with only 11% of respondents listing AI as the top priority for digital investments. Essentially, manufacturing companies are aware of the future potential of AI but recognise that getting there is not going to be easy.
To make AI profitable for business cases, progress is essential needed in some key areas.
Firstly, the date needs to be accessible and usable. Sophisticated AI use cases, such as predictive maintenance or production increases, primarily require a solid data foundation. Therefore, organisations must first take care of the basics and free their business-critical and operational data from silos.
All too often, valuable data from different business units is still not linked or is stored away in disconnected tools. In addition, incomplete, outdated and inconsistently formatted data is a real problem as it cannot be reliably used for AI applications.
Connectivity vs security
Another challenge is data availability. Predictive maintenance, for example, requires a high amount of data connection – even beyond company boundaries. Client companies in the manufacturing industry must be prepared to grant machine manufacturers access to the data generated in production. Despite the overall benefits that ultimately result from this, there are often concerns around exposing sensitive company data.
The solution? Manufacturers should start to invest in company-wide data projects if they have not already done so. Initiatives focusing on connecting data can have a positive impact in many ways beyond AI projects. For example, they can provide businesses with a more holistic view of their own customers and aid in the development of B2B customer portals.
As far as the availability of high-quality operational machine data is concerned, manufacturers should think more holistically and analyse precisely in which cases the transfer of data to machine manufacturers bring more benefits than risks.
The manufacturing industry’s traditionally risk-averse nature is now threatening to become its biggest obstacle. The industry has a zero-tolerance policy for downtime and a strong awareness of the need to protect trade secrets, both for good reasons. However, this also creates a natural hesitancy when it comes to projects involving AI, as implementation often requires wider organisational change.
The solution is simple. By beginning with less risky “simple” AI projects, organisations can gain experience with the management of AI projects. This offers the opportunity to get started and transfer the knowledge to other projects later. At the same time, manufacturers should start experimenting with production data in sandbox environments, especially using ML models for promising use cases such as predictive maintenance.
Organisational change
Many manufacturing companies in the UK are established in an ecosystem that has long since matured. Disruptors that turn the industry upside down from one day to the next remain the exception. There are a lack of role models and best practices, so manufacturers need to catch-up when it comes to innovation and accelerate projects.
Digital transformation is first and foremost a question of organisational change. The existing lighthouse projects relating to digitalisation and AI must be bundled into holistic, centrally managed initiatives. Clear governance structures must be established, and the individual projects must be set up in such a way that synergies are created for the entire organisation. This requires clearly defined AI roadmaps, a strategy for recruiting and retaining specialists and an IT setup that enables KPIs to be measured and data to be exchanged easily.
Change of this scale needs to be driven from the top down and the C-suite should be well informed on how a data-driven approach leads to new, innovative business models. Ultimately, data is just as much a product as the components that are being manufactured and so it should be treated as such.
AI use cases for B2B manufacturing
Bringing order to the documentation chaos An easy-to-implement use case for AI in manufacturing organisations is the use of large language models (LLMs) to search through documentation. These can easily amount to hundreds of pages, and it is often not easy to access the required information, especially under time pressure. B2B manufacturing companies can set themselves apart from the competition by integrating generative AI functions into their customer portals. This enables clients to get the information they need quickly and with low effort.
Predictive maintenance The use of AI in predictive maintenance promises great efficiency gains. However, it also requires a high degree of digital maturity which is not always a reality for manufacturers. For example, high quality and highly available connected data is required as well as collaboration across departments and companies and sophisticated IT management that requires the best talent.
In terms of technical implementation, sensors and cameras are used to provide a wide range of data and continuously monitor the individual components of machines. This can help to predict the most likely down time for machines and calculate the exact moment a component should be replaced, with the least impact on production. Predictive maintenance therefore opens the way for new business models for an entire ecosystem of interconnected b2b manufacturing companies.
The future of AI in manufacturing
AI presents a transformative opportunity for manufacturers but success hinges on the ability to implement organisational change. For AI to become a core priority, manufacturing leaders need to focus on establishing a central strategy. In an industry with varying levels of digital maturity, there are suitable AI use cases for every business where IT leaders can gain valuable experience which can then be transferred to future, more ambitious projects.
Ultimately, long-term success requires a sophisticated data strategy along with organisational alignment. With this in place, manufacturers have the power to turn their AI ambitions into reality and unlock new sources of revenue and efficiency gains through AI business models.
