How to get prepared to adopt Artificial Intelligence in your SME?

For some, it is a revolution that will transform the manufacturing sector, like Fordism did in the beginning of the 20th century. For other, it is an evolution that involves many challenges, but Artificial intelligence will disrupt those enterprises that are ready for it. Let’s see what is to come.

Today the idea of Artificial intelligence (AI) is almost an old chestnut, used to boost webpages’ audience, or to imagine a future with dematerialized devices and robots taking control of our society. Often, it is forgotten that AI is first an intent to simulate and reproduce the human intelligence. Composed mostly of automated processes that are easily identified, but there are also more complex ones which use a huge amount of data and with the final purpose to extend human abilities. When humans are in the loop, there are two kinds of IA: a) the assisted and b) the augmented intelligence, which basically assist and augments human decision-making. In the case of complete separation of human intervention, there is the automation of manual and cognitive tasks and autonomous intelligence that gets adapted to new situations.

Over the past years, we accumulated so much digitalized information that time has come for every sector and especially for the SMEs to take advantages of its benefits. Before crossing the open door and entering a world of intelligence automation, internet of things, machine learning, digital twins etc., it is necessary to consider some important questions, such as: How to get quality data? How much implication of human-workforce is needed for this new technological playground?

What are we playing with?

On the top 3 barriers for AI adoption, according to a survey of 60 manufacturing executives held by Manufacturers Alliance for Productivity and Innovation (MAPI):

• More than half of the manufacturer (58%) invoke a lack of data resources needed to enable AI solutions,
• 52% are uncertainty about how to implement AI solutions to solve specific challenges
• And less than a half (48%) recognize that a lack of sufficient workforce with digital skills to develop and/or implement AI solutions.

Nowadays, data is worth everything and nothing at the same time, but when you want to stay competitive in factory automation, there is a need to ensure the right data and highly skilled experts who will help you activate the ultimate transformation. The gathering of high-quality data and the recruiting of high-level experts is a must to meet successful data analysis processes.

One of the biggest concerns but also the one that is always used when a new technology disrupted the actual model of doing, is human capital. It is true that there is a lack of trained workers capable of managing and maintaining the chore of the software or programmes used to optimize the manufacturing processes. The workforce must be empowered, but also the public must evolve an awareness about AI:

“In order to avoid people becoming overwhelmed by machines, everyone needs to be more prepared for these new technologies and challenges”. (The 2020 World Manufacturing Report).

The solutions to transform SMEs: Kitt4Sme, DigitBrain, Change2Twin

There are many improvements that can be made by using IA related to optimization that allows understanding the demand of clients, such as: having a better vision of the inventory levels, adjusting the production, improving safety in the shop floor, but also allowing product development, improving the process and quality of manufactured products, predicting maintenance and efficient energy management. Medium-sized companies cannot be the only ones to access this technology, scale-ups must also benefit of it.

The European project Kitt4Sme, coordinated by the University of Applied Sciences and Arts of Italian Switzerland, specifically targets European SMEs and mid-caps to provide them with scope-tailored and industry-ready hardware, software and organizational kits, delivered as a modularly customizable digital platform, that seamlessly introduce artificial intelligence in their production systems. This new method would accumulate data from companies, such as information on manufacturing processes, machinery and workers, until developing a system with self-learning capacity that would locate weak points in the manufacturing process and optimize its operation.

Manufacturers that have adopted AI and machine learning to reduce equipment downtime, detect production flaws, increase their supply chain, and abbreviate design times. According to the Capgemini Research Institute, 29% of AI implementations are for maintaining machinery and production assets.

On the other hand, the European program Digitbrain uses AI to prevent machine failures and predict machine needs, giving SMEs easy access to a virtual representation of a product called digital twins:

“The integration of simulation tools, big data analytics, and artificial intelligence mechanisms will boost the operation and performance of the Digital Brain, bringing an innovative solution to rapidly and efficiently model digital twins”

Change2Twin is another European project which supports manufacturing SMEs in their digitalization process by providing Digital Twin solutions too.

How to engage with sustainable AI?

Successfully integrating AI in the manufacturing sector, according to the PWC report, involves three kinds of changes. At a business level, manufacturers need to focus on analytics objectives. From a technical point of view, companies have to work with the right data, systems and tools. Finally, the implementation cannot succeed without identifying the right skills and structure within the enterprise. The democratization and adoption of a data culture, collaboration and agile development is needed as an optimal environment in the factory.

To reach those changes, following the Information Technology and Innovation Foundation (ITIF) there are several recommendations

1. Create teams to drive digital transformation in the enterprise.
2. Define an “AI governing coalition” for AI transformation.
3. Evaluate AI and workforce transformation readiness.
4. Set measurable objectives for digital and AI transformation.
5. Redefine digital and physical product innovation processes. Overinvest in communication for change management.

Artificial intelligence solutions can improve many aspects of SMEs with a good implementation, for that the SMEs must first face some consistent obstacles that could be overcome with solutions provided by the Innovation Actions and several other European Projects within the I4MS community

Marjorie Grassler, In-house consultant at Mobile World Capital Barcelona