Complex Software Landscape in the Automotive Industry – Be Honest!

Uncertainties in the automotive market are increasing the pressure to quickly introduce new software solutions and make data consistently available. However, the IT landscape is often too complex for rapid solutions. Realistic expectations, precise business cases and planning with time buffers can be successful.

Few had anticipated this: A couple of months ago, a software update knocked out IT systems at airports, hospitals, and banks. Transactions were impossible for several hours. The issue highlights the fact that software and its interfaces in companies have become complex and sometimes unclear – across all industries. 

In the automotive industry, where software is becoming increasingly important as a key technology for developing new vehicles and business models, the pressure to quickly implement new solutions and make data consistently available is growing. The consequences of improper integration of software components can be very serious.  

If one “gear doesn’t run smoothly,” the whole machine suddenly stops. Or if a line in the software code is written incorrectly, unexpected errors occur, so-called bugs. How does this happen?

One answer is: the software landscape is a complex system, and the realization of new IT projects in companies is too often rushed or based on inadequate requirements. Investments must pay off quickly, which can put CIOs and digital transformation leaders under pressure. 

Set up projects properly and with a time buffer

The more crucial it is for companies to be honest with themselves. This begins with admitting that software in the company can be complex, and when planning new projects, time buffers should ideally be incorporated from the start. Put simply: if you do something too quickly, you do it twice.

A clean approach to introduce new software starts with:

• Clearly naming and prioritizing expectations and not just state them – in other words, instead of doing everything at once taking a step-by-step approach 
• Providing the necessary specialists depending on the implementation phase. 
• Seeking advice from experienced service providers to ensure you know what is needed. 
• Carefully reviewing and simulating existing and future processes through modeling before they get digitalized. 

The perspective of a “super-architect” is not enough here. Experts should consider the new processes together from various angles – including enterprise, domain, solution, and system architects, the relevant business departments, and the employees working with the machines and screens.

Added value through consistent data: Sharing design ideas with factories 

The basics of project management also include regular monitoring of technical requirements and capacities, user training and meaningful escalation if milestones are not reached on time. All of this ensures that added value and confidence can be generated from modern software. Having data available at all times is certainly one of the most important success factors for companies. At best, new developments are already available digitally and in a usable format on the product, machine, process, and application side, so that all steps from production to product transportation can be tracked and digital consistency is achieved.

Artificial intelligence optimizes production processes

Advantage: A construction idea in 3D can be digitally exchanged with factories and checked for manufacturability. AI now converts digital parts lists from development into production bills of materials, enabling faster preparation for manufacturing. Production processes can be analyzed by AI or generative AI (Gen AI), optimized via process simulations, and products can finally be provided to market much faster. AI is also used when thermal cameras automatically check the temperature of batteries and monitor them 24 hours a day. Even data glasses in production help assemble components, but all of this requires continuous data on robust networks. However, this is not always a reality, especially in production, where many systems are several years old. Moreover, individual systems were often developed on-site with software companies or internal employees. Besides the lack of documentation and employees who are no longer with the company, these systems often cannot communicate with modern architectures without significant technical effort. This can disrupt the overall IT architecture, resulting in security risks. A possible way out is to use open, platform- and cloud-based solutions that can be combined with local computing power (hybrid). Standardized data or interfaces and fewer applications can help reduce complexity and costs across the company while also creating transparency worldwide.

Business case with realistic expectations 

“Being honest” also means conducting interim assessments. How many people are dedicated solely to maintaining legacy systems, how many are involved in project initiatives, how much money is spent on licenses, and where is software with real value being used? A comprehensive strategy is not only based on well-designed IT architecture and good project management but also on a business case that is realistic and does not create false or inflated expectations. In collaboration with experienced partners, the goals should be described with clear assumptions and measurable success factors. This is especially relevant for companies operating in a competitive market environment, such as the automotive industry. A modern, highly flexible vehicle factory from an automaker offers a good example. It relies on a modern IT infrastructure, advanced software solutions, hybrid data management, and software views that are easy for employees to use. This enables the factory to be flexibly adapted to changing production needs. Additionally, the performance indicators currently collected can be used to control autonomous robots in warehouses and manufacturing, allowing the company to respond to market trends and remain successful in the long term.

Of course, risks to an automotive manufacturer’s production line do not only come from the software landscape. However, these risks can usually be reduced or even avoided with clear expectations, measurable goals, well designed processes, suitable software, and extensive pre-testing.