Comment: Four forces driving intelligent manufacturing

Manufacturing companies that choose to leverage data and AI will benefit from a business model that is smarter, more flexible and more competitive than ever before, says Bala Amavasai, global technical director, Databricks.


The slow natural pace of manufacturing’s evolution used to be over decades. However, recent years have turned this natural pace upside down. Dramatic falls and spikes in demand came with repeated global lockdowns, massively disrupting supply chains. What did this turbulence teach manufacturers? That digital is key. In the UK, for instance, 80 per cent of manufacturers believe industrial digital technologies will be present in their business by 2025.

This rush towards digitalisation is making way for “Intelligent manufacturing”. This makes use of Internet of Things (IoT), cloud computing, data analytics and machine learning (ML) to optimise how organisations use assets, with a focus on the return of invested capital (ROIC). Amid all this change, there are four main challenges facing manufacturers which, if harnessed correctly, could also become driving forces of change. 

Skills and production gaps

Intelligent manufacturing’s use cases are placing high demand on robotics programmers and technicians, cybersecurity experts, digital twin architects, supply network analysts, and people who can leverage data science and ML algorithms. This means the industry is facing challenges around training and retaining staff. In the UK, according to the latest ONS figures, there were 95,000 manufacturing vacancies in April-June 2022 – a 49.5 per cent year-on-year increase. 

One way of addressing this challenge is to upskill and reskill existing workers in emerging technologies, such as collaborative systems and advanced automation tools. Many employees may have existing technical skills that can be bridged into these new use cases. Open source technology could also provide support for manufacturers needing skills in the short term, particularly skills around data. Open source allows organisations to tap into a wider set of skills and expertise, harnessing the power of peer-to-peer learning and alleviating pressures on existing teams. 

Supply chain volatility 

If the effects of the pandemic proved anything, it’s that supply chains need to be robust, transparent and resilient. The ability to monitor, predict and respond to external factors – including natural disasters, material shortages, and shipping and warehouse constraints – is vital to reduce risk and promote agility. This requires up-to-the-minute, end-to-end visibility across all stages of the supply chain. However, many manufacturers currently operate on complex legacy data architectures, like data warehouses, which can cause information silos to form – preventing easy data access and distribution. Inaccurate datasets that contain duplicated or outdated information may also be inadvertently shared. 

The answer lies in building a strong, modern data foundation – such as a data lakehouse that reduces the number of platforms needed – removing complexity and enabling greater visibility and access to data. By ensuring the timely flow of accurate data, as well as AI and ML use cases, a modern architecture is primed for real-time analysis and insights. This allows manufacturers to make key decisions as and when they receive insights, responding quickly to change and therefore riding out volatility far better. 

Need for new sources of revenue

Manufacturers’ growth has historically been limited to new product introductions or expansion into new geographies. The emergence of equipment-as-a-service (EaaS), however, is changing that dynamic. While this approach is not new (Rolls-Royce’s “Power-by-the-Hour” engine subscription model has been around since 1962), customer demand, advances in industrial IoT, and a continuing decline in sales and margins have seen EaaS emerge as imperative for manufacturers. Why? Because it offers a level of visibility and collaboration that requires lower maintenance costs, capital expenditure and human capital management. 

We can look to Rolls-Royce as an example. It has used cloud-based models to reduce costs for customers and to create new revenue streams. Rolls-Royce has been able to collect real-time data generated via the creation of digital twins of its engines. This data is then analysed with AI and ML to help avoid unplanned grounded planes, and reduce millions of pounds from inventory parts costs. 

A need for greater sustainability

The impact of climate change has led to growing volatility in global supply chains. As such, manufacturers must do more than adapt, and must be active in reducing the industry’s environmental impact. Manufacturers must also consider the indirect emissions resulting from activities outside of their control. 

Achieving greater sustainability requires establishing a redesigned, circular supply chain. It will require greater levels of collaboration between suppliers and vendors, the optimisation of production lines and transportation, and greater levels of customer engagement to extend product lifecycles. This all starts with data. Collecting data into one place makes it simple to access, as well as storing it for analysis and AI and ML use cases. This provides visibility and intelligence across the entire network, allowing manufacturers to make key decisions to improve efficiency. Employing open-source software will also be key here, enabling greater data sharing and collaboration across manufacturers, suppliers and vendors up and down the entire supply chain. 

In today's world of work, it’s data-and AI-driven businesses that are being rewarded. Manufacturing companies that choose to leverage data and AI will benefit from a business model that is smarter, more flexible and more competitive than ever before.

Bala Amavasai, global technical director, Databricks