.webp)
Software analysis and testing play a crucial role in industrial automation. While manual testing remains vital for addressing nuanced issues, especially those related to human safety, automated testing offers distinct advantages in terms of speed, scalability, and cost-effectiveness, ensuring the reliability of automated systems.
Integrating automated analysis and testing protocols empowers the industry to boost productivity and enhance product quality while mitigating the risks associated with human errors, which can be time-consuming and costly to rectify.
Having a robust architecture verification process serves as a cornerstone for refining code in order to make automation effective and compliant. It helps streamline workflows, improve data flow, and enhance overall performance by identifying deviations in code compliance, potential bottlenecks and vulnerabilities.
Closing the Innovation Gap with Architecture Verification and Static Code Analysis
A McKinsey report highlights that manufacturers often find themselves in an ‘analysis paralysis’ before rolling out Industry 4.0 solutions. They miss out on the shorter time-to-impact enabled by a proven and pragmatic minimal viable architecture when they wait for a fully-fledged, ideal-state data and IT/OT (information technology/operational technology) architecture to be defined and implemented.
“Interoperability and integration are key considerations. Automation systems should seamlessly integrate with various IT systems to facilitate data transfer, transposition, and orchestration,” says Prashanth H. Southekal, founder and managing principal of DBP Institute, a data and analytics consulting, research, and education firm.
It is a sentiment echoed by Dr Daniel Simon, Quality Assurance Customer Service at Qt Group, who emphasizes the importance of maintaining software architecture: “If you don’t keep track, you lose your overview of complexity and encounter trouble. You are liable for bad time-to-market because the software erodes over time and becomes less maintainable. We call it the ‘innovation gap’ - at some point, you reach a level of complexity in your software that you cannot handle. It cannot be changed anymore. That is a risk; new systems will be brought to the market by your competitors, and you’re left behind.”
Architecture verification and static code analysis address gaps due to architecture and software erosion and amplify control to manage complexity in advanced systems. Combining architecture verification with other code analysis features, such as guideline checks and clone detection and management early on, can greatly diminish the time and expenses associated with rectifying style and coding violations later in the development process. Simon explains that along with verification, timely updates are essential to keep systems running smoothly, as software bugs are inevitable. “Together, they can help manufacturers manage their overall software lifecycle.”
Complimenting Static Analysis with Dynamic Testing and Code Coverage
When static code analysis is combined with dynamic testing and code coverage, testing strategies evolve into powerful tools for scrutinizing the system's runtime and delivering quantifiable data. As Jan Aarsaether, Senior Business Development Lead, Quality Assurance Tools at Qt Group, explains, “Code coverage determines what specific parts of the source code your tests cover and, more importantly, what they do not.”
It enables efficiency in testing strategies, optimizing resource allocation, eliminating false positives and maximizing testing effectiveness.
“In the end, it’s about testing smarter, not just harder. With more machines and the agile DevOps way of working, release cycles are shrinking and becoming competitive, and the problem is escalating. If you don’t work smarter, you won’t be able to do it.”
Common challenges include integrating richer applications that run on a variety of hardware, dealing with multiple operating systems and screens provided by different hardware and platform vendors, and always being release-ready due to Agile DevOps-style technologies and processes.
The Power of GUI Testing
Increased digitization and process automation have led to a proliferation of both visible and invisible applications, leading to a surge in the number of screens in the industry. These screens, which were previously mechanical buttons, have now become monitoring interfaces for sensors or robots that workers interact with to control various aspects of operations.
Introducing new capabilities means introducing new software behind these interfaces, such as expanding the Graphical User Interface (GUI) for monitoring or controlling production processes. This expansion increases software and screen diversity, amplifying the scope of work needed to ensure system efficiency and reliability.
Aarsaether says, “A majority of those doing automated testing have adopted job methodologies. However, functional testing also needs to be automated because the scope of the problem, not just the amount of work, is increasing manifoldly. Manual testing, which is the alternative to automated functional testing, is no longer feasible since you will find issues very late in the process when they are both more difficult and expensive to fix.”
Prioritizing Regulatory Compliance
Approaches such as static code analysis, code coverage analysis, and GUI testing help not only make businesses more productive but also assess and ensure regulatory and standard compliance.
These processes jointly provide documented proof of compliance with regulatory requirements, such as test reports and traceability matrices, which are essential for audits and investigations. They also help manufacturers validate software systems, adeptly manage changes and bolster continuous improvement initiatives.
“Regulatory approaches, particularly concerning the safety of software systems that could pose risks to life or health, are well-established. For instance, ISO standard 26262 applies in the automotive field; in the medical field, it’s ISO 62304. In general, ISO 61508 serves as a guideline, emphasizing the importance of ensuring that software functions properly. It also stresses the need for careful planning before implementation, a step often overlooked,” says Simon.
The Way Forward
Industry experts say that the need for verification, testing, and regulatory compliance will increase as we advance. A 2024 Gartner industry-wide survey of 2,400 CIOs across 84 countries found that improving operating margins (65%), generating revenues (46%), ensuring compliance, and minimizing risk (45%) were the most critical outcomes for digital technology investments.
“I think what we’re going to see going from Industry 4.0 to 5.0 is a focus on increased granularity. But that also brings with it a challenge, which is managing the complexity. It’s going to be much more difficult going forward,” predicts Peter Aiken, Associate Professor at Virginia Commonwealth University.
“We also need to acknowledge the fact that, of course, humans program all of these. Humans make mistakes, and therefore, the software makes mistakes.”
Experts unanimously agree that human oversight in industrial automation is critical to achieving seamless operations. This is particularly true in mixed environments where human safety is the main priority and humans and robots are coworkers. They conclude that allowing humans to focus on what they do best and automating repetitive tasks is the ideal way forward.
