Fixing the system complexity problem in industrial automation
By Michele Rossi
As digital transformation continues to proliferate, software development processes for industrial applications must evolve as well.
By Michele Rossi
While the industrial market is fragmented into thousands of different application areas, the overall focus on automation has grown dramatically in recent years – as have the demands for remote accessibility, connectivity and the like. By 2025, automated machines will perform more current work tasks than humans, according to the World Economic Forum.
The rapid evolution of machines and algorithms have resulted in a wide variety of benefits for industrial organizations looking to improve operational processes. Yet emerging strategies, such as Industry 4.0 and the Internet of Things (IoT), have also introduced new challenges to factories deploying sophisticated, modernized systems. Examples include the relationship between the volume, velocity and variety of data and the complexity of software applications to power these contemporary devices.
Indeed, connected devices are processing and analyzing growing amounts of data – from an increasing number of industrial sensors – and are measuring a more extensive range of data types (moving beyond video and audio, to, for example, vibrations and acoustics). As a result, the software running this equipment has become harder to develop, deploy and manage. To ensure organizations implement effective industrial automation – and receive optimal ROI – current technical and operational practices must adapt quickly, efficiently and successfully, including the state of device infrastructure, application complexity and development.
Following is an examination of modernized software development technologies and practices that industrial automation manufacturers can leverage to deliver innovative, user-friendly solutions as customers’ needs grow. This analysis will be followed by insights into how these software standardization initiatives work in practice.
Industrial organizations are undergoing widespread digitalization and automatization – what are the resulting challenges?
The focus of automation has expanded from increasing productivity and reducing costs to also improving quality and flexibility in the manufacturing process. For example, frontline staff can train modernized, automated equipment to perform tasks previously thought to be too difficult for standard machines, like picking and packing irregularly spaced objects or resolving wiring conflicts in large-scale projects.
To keep pace with end-user demands for highly innovative solutions driving Industry 4.0, organizations are implementing advanced automated technologies designed to handle a wide range of processes and machinery. According to the McKinsey report “Human + machine: A new era of automation in manufacturing,” the new generation of machines are more flexible, versatile and cost-effective compared to devices used in many industrial environments today.
While the pressure to automate isn’t unique to a single vertical, very few industries must deal with the sheer volume and complexity of devices, systems and users that industrial organizations face – which are ever-evolving.
Although industrial applications can leverage many things from mainline application development, they often have more specific requirements than the average web portal or mobile or desktop app:
- Industrial applications must be reliable; mistakes can be terrifically costly in terms of destroyed material, ecological consequences or human endangerment
- The need to monitor or control applications remotely – via an operator’s tablet, smartphone or desktop – is a general requirement for industrial applications
- Industrial applications are often both resource-constrained and performance-critical. The software needs to be able to communicate directly with devices or protocols, demanding precision timing
- While all software should be secure from intrusion, industrial applications demand it due to the sensitivity of the data they handle
Some think software development for industrial applications hasn’t changed much in the couple of decades. While it’s true that the space is more resilient to the whims of consumer desire, a proliferation of new requirements means that industrial organizations must evolve their development processes accordingly. These new requirements include support for new protocols; needs for visualization and charts; customer demand for internationalization; touch screens and virtual keyboards; support for headless devices; and easily-interpreted dashboards.
For example, a user of an industrial device that measures voltage levels expects a modern, interactive and easy-to-understand display to visualize power levels. But integrating all the various pieces of software required to make an automated machine run successfully can be prohibitively complicated, costly and time-consuming.
Lastly, numerous market trends have affected the look and feel of industrial applications and how they are designed and built. The adoption of handheld mobile devices changed the worldview of what an acceptable user interface should look like – even for industrial applications. The convenience that smartphones and tablets provide has made a permanent impact on customer expectations about ease of use, functionality and immediacy.
To produce increasingly complex, automated systems, industrial automation manufacturers must optimize software development processes to boost time-to-market, improve product reliability and meet customer expectations.
Standardizing across software
To improve product development efficiencies, industrial automation manufacturers can implement tools, resources and techniques to develop reliable and effective products faster. Device creators could expedite development processes by using tools and technologies available on the market to streamline the iterative prototyping process, like screen generators, design tools and software factories.
For example, if an organization that creates PLCs (programmable logic controllers) and industrial touch panels delivers devices with software tooling already included, system integrators can modify parameters and create animations on the touch panel quicker and more efficiently – which significantly reduces the time-to-market of these devices. By making software development more efficient, device creators will reduce the development cycle and instead focus on safety, reliability and efficiency – as well as time-to-market – allowing manufacturers to realize new product revenue sooner.
Another significant part of reducing development cycles is in improving the workflow between designers and developers. Currently, the software development processes of industrial devices include both designers and developers – separate people with different specialties and levels of expertise. The designers create the look and feel in the form of a mock-up first, which is then handed off to translate into an actual thing by the developers. By alleviating the sometimes-painful round-trip between designers and developers caused by different, incompatible tools, organizations can reduce the time-to-market for industrial automation devices.
A prime hypothetical example is an industrial welding machine manufacturer operating distinct teams across the globe for front-end and back-end development. If this manufacturer chooses to adopt more efficient software development processes, including tools to enhance the designer-developer workflow, the manufacturer can reduce project and communication complexity – thus improving time-to-market of its devices.
By investing in technologies that can make the software development process more efficient, streamlined and cost-effective – without having to continually rewrite code – industrial automation manufacturers can innovate at scale and deliver on their customers’ demands.
Use case example
Industrial gateways, touch panels, mobile applications, web applications and engineering tools are aimed to optimize time-spend by users in-the-field and deliver more effective results. Thus, it is crucial that software applications provide excellent user experiences and have the flexibility and portability to run successfully on-top of any IoT-connected or automated device. Without adequate, modernized software development processes, organizations can experience delayed time-to-market or deliver ineffective, hard-to-use devices to its customers.
For example: a manufacturer that creates industrial touch panels for the transportation industry discovers that its customers are struggling with the complexity and usability of its in-cabin displays. The manufacturer had been looking for new ways to increase customer value and support growth, and as a result, it wanted to reduce the number of displays in the machinery cabin – which required the integration of several applications into a single display.
Because the manufacturer’s systems required a proliferation of displays for several disparate systems, it was not only more difficult for users to operate, it was also more expensive to manufacture. To create more user-friendly interfaces and bring machines to market quicker, the manufacturer implemented more efficient software deployment tools and processes, which allowed the company to simply and quickly assemble custom applications from a large assortment of pre-built and tested components.
The overall focus on industrial IoT technologies has grown dramatically in recent years, as have the demands for different deployment scenarios, remote accessibility, connectivity and consumer-grade interfaces. As digital transformation across industrial organizations continues to proliferate, software development processes for industrial applications must evolve as well – so that industrial organizations will be able to addresses critical trends and challenges and deliver increasingly complex systems that maximize production uptime and product reliability.
Michele Rossi is product manager, automation at Qt.