Manufacturing Predictions - the highlights for 2023

After having accelerated cloud and digital transformations during the pandemic, manufacturers might be forgiven for wanting to take a pit stop. But there is no time to slow down and rest on their laurels. On top of inflation, energy price shocks and economic uncertainty the technology landscape is shifting, faster than ever before — and so are the new kids on the block such as massive cyberattacks and sustainability imperatives. 

Let's delve into two key manufacturing trends for 2023 and keep ourselves honest about a very familiar buzzword.

1. Security posture modernizations, in particular on OT Security
2. Sustainability commitments, in particular reduction of the CO2 footprint
3. AI - Where are you?

Security posture modernizations, in particular on OT Security

Manufacturing is the most targeted industry for cyberattacks. Industrial organizations expect increasing cyberattacks on connected devices, especially legacy control systems. Organizations are reeling from increasingly sophisticated attacks with hefty ransomware demands and, to make things worse, there is an acute security talent shortage. NIS2 is coming and according to a large Swiss insurance company, cyberattacks will become “uninsurable” due to the unpredictable risks and costs involved. All in all, not exactly a relaxing outlook for CISOs.

Operational Technology (OT) Security had not been on the radar of many CISOs and plant managers until they got hit. To be realistic, in the past there was very often no need to protect production plants and machinery from a cyberattack, simply because there was no connection to the internet.

Nowadays, the increasing digitalization, IIoT and the convergence of IT and OT are creating new security risks - in particular when taking into account that the market penetration of cybersecurity in OT markets is very low.

Legacy systems and increasingly interconnected devices expose the vulnerability of manufacturers – in many cases they are “naked” and easy victims. Industrial control systems (ICS) with their key significance for manufacturing operations are often more than 20 years old and as such a sitting target for cybercriminals once getting directly or indirectly connected to the internet. Even so-called “air-gapped environments” are no longer safe since there is always a risk by ever-present Wi-Fi connections in the plants.

After reacting in a defensive mode for the past years, manufacturers now realize it is time to step up with a proactive security strategy including OT.

So, what to actually do? 

The above-mentioned challenges are fueling security modernization and resilience. Many manufacturers are looking at vendor consolidation, investing in interoperable security tools, and implementing automation & orchestration technologies across their IT and OT environments. Needless to say, a specific focus on integrating OT Security into an IT / OT SOC will be a required step to realize the much needed holistic visibility across the levels 0-5 of the Purdue Model.

Finally, it is no secret that manufacturers used to be very hesitant, even outright rejectful when it came to the question of cloud adoption. However, this opinion has significantly changed, especially in a security context. “Security and Risk Reduction“ were among the top three reasons for cloud technology implementation as highlighted by a McKinsey study.

Increasingly, organizations are also deploying their Security Information Event Management (SIEM) in the cloud (aaS) simply because of the many benefits such as the flexibility of a hybrid architecture with automatic software updates, agility in an ever-changing business landscape and innovation.

Ultimately, it will be a journey by starting with foundational visibility when including monitoring firewall data of all OT locations up to optimized experiences when leveraging sophisticated tools to avoid fighting large volumes of noisy alerts.

Sustainability commitments, in particular reduction of the CO2 footprint

The World Economic Forum has picked up sustainability as a key theme, supporting multiple industry initiatives such as 50 Climate Leaders, which not surprisingly, many of them are manufacturing companies. 

In fact, the manufacturing sector bears a particular responsibility with a 22% share of Europe's total CO2 emissions and the share in the US is similar. 

Driving the movement towards increased sustainability, the EU has adopted mandatory Corporate Sustainability Reporting starting in 2024. Also, a good sustainability standing with an acceptable CO2 footprint will be an important source of long-term competitive advantage: Millennials and Gen Z are the driving growth towards more sustainability which manufacturers must take into account. With Europe's carbon price hovering between 70 and 100 EUR/t there is obviously also a cost savings factor to it as well.

However, the challenge is the lack of transparency into the CO2 footprint. Organizations lack the much needed holistic visibility of the CO2 footprint across cloud, hybrid and on-premise environments - across scopes 1-3. Businesses struggle to measure and monitor something they know is so important. The old management wisdom “You can’t manage what you can’t measure” applies. 

Only a few leaders can answer the simple questions with one number: What is your CO2 footprint? 

So, what to actually do?

The link between digital transformation and sustainability is often overlooked. However, industry leaders proactively embed digital technologies to enable a data-driven sustainability strategy.

Most importantly, each organization is on a different stage of such a journey. Many manufacturing companies can start with a “Green IT”. Cloud migration of on-premise data centers is by far (!) the easiest and most effective way to start or accelerate a sustainability journey since the CO2 footprint can be reduced by 80% on average by such an exercise. Also, data center optimization with real-time and predictive insights of the CO2 emissions is a classic low hanging fruit which should be harvested in a first step.

On the other hand, manufacturers which have moved to the cloud already can go ALL IN, meaning leveraging data analytics for holistic visibility across scopes 1-3 and across cloud, hybrid and on-premise environments.

Ultimately, same like with OT Security, it will be a journey by starting with lowering the CO2 footprint with “Green IT” over manufacturing specific use cases and tackling the supply chain. The ultimate step is going ALL IN with holistic visibility across scope 1-3.

AI - Where are you?

The Non-Millennials among you may recall the famous Science Fiction movie “2001: A Space Odyssey” where the supercomputer HAL predicts a malfunction which turns out to be a mistake. Before being switched off it develops vicious resistance and becomes very emotional.

Nowadays, there is a running joke of the factory of the future: The factory of the future has only one worker and a dog. The worker’s job is to feed the dog and the dog’s job is to keep the worker away from the machines.

So, where is the AI now? There is a significant difference between “Narrow AI” and “Advanced AI”. Narrow AI relies on computational powers and automation based on a framework with “certain” parameters - but is this really AI? In contrast, Advanced AI flourishes in unknown and “uncertain” situations without any human input - the key ingredient of blockbuster movies. However, if AI “knows it all” is there a need “to predict anything”?

Needless to say, manufacturing companies will welcome the benefits of AI resulting in higher productivity, higher availability, higher product quality, lower cost and most importantly new business models which will make both shareholders and customers happy. 

However, the bitter truth is that AI will still remain intangible for many manufacturing companies  because of the realities on the shopfloor.

So,... what are you going to do about AI?

Most importantly, it’s not about complex statistical gimmicks and algorithms but about concrete value Narrow AI can provide in the real industrial world. Application examples include quality control and automation of repetitive work with the help of robotics. One very beneficial service application will be for example the automation of warranty claims processes. One promising option to implement Narrow AI is to leverage external Machine Learning expertise whilst providing the required domain expertise.

Conclusion

Predictions are a funny thing. “I think there is a world market for maybe five computers” Thomas Watson, chairman of IBM famously predicted in 1943. And Paul the Octopus’ accurate predictions in the 2010 World Cup brought him worldwide attention as an animal oracle.

Next year latest we will know if security posture modernizations and sustainability commitments have become true and if AI has finally been broadly implemented.