Skip to main content

How could AI and automation tackle the UK's collapse in car manufacturing?

artificial intelligence
(Image credit: Image source: Shutterstock/PHOTOCREO Michal Bednarek)

1. What are the main factors contributing to the record fall in UK car manufacturing levels?  

The U.K. automotive industry has been a pinnacle of excellence over the last century. However, during the last few decades, sectoral shifts and an evolving competitive landscape have adversely affected the industry, with the pandemic further aggravating these challenges by throwing the demand-supply equilibrium into disarray. 

The recent and historic fall in car manufacturing in July – which saw production fall to its lowest level since 1956 - is a combination of factors. In an industry as resource intensive as car manufacturing, the success of every manufacturer hinges on how well they navigate both local and global market challenges, such as staffing and material shortages. On one hand, the ‘pingdemic’ has meant that carmakers have had to deal with unexpected staff shortages at a local level. More globally, the rising prominence of semiconductors in today’s tech-powered products have meant that if manufacturers can’t cope with an ongoing microchip shortage, production often comes to a grinding halt. That said, I am confident that the industry is resilient and will bounce back to normalcy, stronger than ever, and expect technology will play a key role in enabling this resilience in a prominent way.

2. In what ways could technology have been used to predict or mitigate the UK’s fall in production levels?  

Many of the challenges faced by manufacturers over the past year were unprecedented, but it’s important to note the role that technology could have played to predict or mitigate these challenges. For example, a transparent supply chain that helps to spot constraints in global shipments could enable early identification of supply chain bottlenecks and help inform the steps needed to clear them, or plan for alternative sources or logistics. For this reason, a smart, connected supply chain, which leverages data analytics to predict imbalances in supply and demand would have been a great way to proactively respond to an ongoing semiconductor shortage before it contributed to a fall in production.

Another contingency plan would be to standardize product configurations as much as possible, and postpone their final assembly, depending on the availability of critical components like chips and customer demand. 

In addition, to counter the possibility of further staffing shortages, manufacturers could consider adopting higher levels of automation. Automation leverages advanced technologies such as artificial intelligence (AI), business process management (BPM), and robotic process automation (RPA) to automate tasks normally carried out by humans. But rather than to replace humans, automation could be deployed across functions that use up a lot of staff time, with little return. By automating these tasks, businesses could improve the efficiency and reliability of these tasks, enabling staff to spend a greater amount of their time on the tasks that drive the biggest results. 

Ultimately, technology is a tool that carmakers can leverage to strengthen their core, drive intuitive decisions, and build responsive value chains. At Infosys, we call it a Live Enterprise, an organization that continuously senses its environment and responds in an agile, innovative, and rational manner. Think of it as an organism that adapts to its natural environment with the help of technology.

3. The manufacturing sector’s digital transformation efforts began long before the pandemic – what’s holding back its progress?  

There are of course a number of challenges holding back the sector’s digital transformation. Industrial productivity has remained flat for a decade, according to McKinsey. Technical debt from legacy systems continues to be a drag on some smaller manufacturers, sometimes impacting their ability to invest in contemporary systems. On-premise enterprise systems are also weighing down many organizations, preventing them from adopting secure collaboration tools for interactions with external partners and from scaling up globally. 

On the other side of the coin, many large organizations are well and truly committed to digital transformation, adopting initiatives like Industry 4.0 or the fourth industrial revolution to drive the uptake of smart technology in operations. However, to avoid rushing the process and contributing towards organizational siloes, it’s important that digital transformation efforts are executed carefully and thoughtfully. For example, AI adoption in manufacturing requires a lot of data, which needs to first be cleaned, structured and stored in a way that ensures security and is compliant with data regulations. 

In addition, the high number of use cases – which all perform at different levels of maturity - makes AI adoption in manufacturing an incredibly complex endeavor. Therefore, the first – and too often overlooked – step for manufacturers is to assess the organization’s AI maturity, which in turn should inform the organizations’ transformation strategy. This process involves critically assessing the organization’s current state of AI adoption and aims to gain consensus on its short- and long-term goals, before advising on the execution of AI integration. Using these insights, organizations’ digital transformation efforts could be planned and implemented to ensure that it is cost-effective, low risk, and delivers a ROI. 

It’s promising to see that initiatives like ‘Made Smarter’ - commissioned by the government - are supporting small and medium enterprises in their Industry 4.0 journey. The four recommendations made by ‘Made Smarter’ are - leadership, adoption, innovation, and skills. Amid a staffing shortage, skills are likely to be the biggest challenge holding back the progress of digital transformation, but I’m optimistic about what the sector could achieve.

4. The UK’s manufacturing sector has long debated the contribution that greater automation could bring to its day-to-day operations – what are the main benefits and challenges?  

The proven benefits of automation are higher levels of accuracy, quality, productivity, and safety. During a staffing shortage, automation also reduces the uncertainty surrounding the labor required to manufacture automobiles. Leveraging AI, automotive tools can make predictions based on the data they receive, enabling better product uniformity, quality, and safety. When it comes to car manufacturing, where mistakes simply cannot happen, automation provides an additional layer of safety for all involved in the product’s lifecycle. Given that these tools could also enable remote operation and monitoring, factories could effectively reduce the need for human-to-human contact during the ongoing pandemic.

However, the widespread take-up of these tools has understandably been beset by a range of cultural, financial, and logistical challenges. The primary challenge in the implementation of automation is the skill set required to implement it, from the design of the system to its installation, commissioning, operation, ongoing maintenance, repair, and upgrades. Change management is ultimately a challenge with a workforce that is used to traditional methods of manufacturing. In these circumstances, talent transformation platforms like Infosys Wingspan could play an important role in ensuring upskilling programs are led and informed by technology. 

Amid a cultural resistance to automation due to the fear of job losses – particularly amidst a pandemic – upskilling initiatives could enable staff to take on more human-centric roles while leaving automation to take care of more mundane, time-consuming tasks. It’s ultimately crucial that organizations are transparent about their automation plans and have a robust change management strategy in place well ahead of time.  

5. How could businesses enable greater automation?  

Traditionally, the focus of automation in the industry has been reliant on using robots and Automatic Guided Vehicles (AGVs). Today however, newer technologies such as machine learning are enabling automation of various Quality Assurance (QA) processes. 

Connected systems are a key prerequisite for deploying automation at scale and therefore any successful automation project must consider an effective integration strategy. This will allow for systems to be managed centrally and communicated throughout an organization. We anticipate that the rollout of 5G will support manage the complexity of this process over the coming years.

In addition, every automation project will need to ensure compliance with regulatory requirements regarding data and AI. The success of any project will depend on selecting the appropriate data for each use case and ensuring its quality and security. At Infosys, we use a simple framework to assess automation possibilities. The complete list of tasks required to make a car are classified into four categories, depending on the amount of programmability and decision making. Jobs that are high on programmability and low on decision making are best suited for automation. These are repeatable tasks with low levels of uncertainty, and so robots in the shop floor should be able to can handle these tasks with ease. Tasks that are low on both factors are jobs that require a human touch. Humans will always perform them. 

Tasks that are high on both programmability and decision making are fit for hyperautomation.

6. What is hyperautomation and what role is it likely to play in the manufacturing sector over the next decade? 

Hyperautomation takes automation one step further and involves the orchestrated use of multiple technologies and tools which work in harmonized fashion to augment repetitive tasks. Beyond AI, these technologies and tools include machine learning, robotic process automation (RPA), business process, management (BPM), intelligent business process management suites (iBPMS), and more. 

There are, understandably, worries that hyperautomation will impact job security in the U.K. over the coming decade. However, no single technology can replace the capability of humans. Humans will continue to play a role alongside machines and algorithms. The pandemic has significantly increased the importance of usage of hyperautomation. It will play a key role in manufacturing in the next decade. But jobs that need high levels of creativity, problem-solving, and decision making will continue to involve humans.

Close to 3 million people are employed in the manufacturing sector in U.K. Interestingly, investments for hyperautomation also take into account the workforce planning required to prepare this talent pool for the future. Initiatives for skill development, such as flexible working, apprenticeships, continuous education programs, etc., should be taken up to support lifelong learning and plug the skills gap. With such initiatives, humans will continue to play a key role in hyperautomation. In my view, when machines and algorithms take up repetitive tasks with high levels of accuracy and precision, people get freed up for new and innovative pursuits.

Jasmeet Singh, Executive Vice President and Global Head of Manufacturing, Infosys

Jasmeet is Executive Vice President and Global Head of Manufacturing at Infosys, a global leader in next-generation digital services and consulting.