There is much talk about the growth of low-code and no-code development platforms. Indeed, Forrester says the low-code market will top $21 billion in spending by 20221 and Gartner forecasts that low-code application platforms will account for 65% of all app development by 20242. With growth rates of over 30%, three times greater than the enterprise applications markets at 10.2%, this means most apps created in 2024 will be developed using platforms and tools that provide turnkey ways to program. So, will low-code or no-code platforms be the future for manufacturing? What pros and cons do they bring to the shop floor environment and where do Manufacturing Execution Systems (MES) fit in?
A low-code or no-code rapid application development (RAD) platform uses a drag and drop visual software development environment to enable users to create a mobile or web app. No-code has the distinction of creating these apps without writing a single line of code. This approach means much faster development and agility compared with writing lines of code, resulting in quicker updates and changes to the system being developed. It reduces the risk of programming errors and removes the need for skilled programmers, which can be difficult to resource. A RAD platform can further improve business governance, facilitate application analytics, and reduce lifecycle support costs. It has also been cited as an integration platform to share best practices while working remotely, something which has been of extreme interest over the past year of lockdowns.
At face value, the appeal of a RAD platform is compelling, offering the ability to easily create purpose-built apps for every process, customized for the specific workforce and application. The problem is, however, that building an app for every operation or process in a manufacturing environment may result in hundreds of discrete, disconnected apps.
The primary challenge with low/no code for manufacturing is that there is no means to create a foundation for the manufacturing process. There is no common plant model, no common set of data definitions, no central database to store and analyze the processes, and no cohesive, plant-wide infrastructure. In the case of manufacturing, it’s not about the app, it’s about the process, and this is where modern MES solutions come into play.
RAD and MES
A key difference between MES and RAD platforms is the execution: both from an implementation and aftermarket support. MES uses an ecosystem of partners and vendors, whereas many RAD platforms argue that no external resources, such as partners or third parties, are required for development or deployment. Is this a good thing?
Modern MES systems now have an advantage over RAD as it already incorporates the benefits of RAD in addition to the comprehensive functionality contributed by the data platform. By selecting an MES with low/no-code functionality, the benefits of extensibility (low code) are reinforced by the stability of a data platform and MES data model.
For older MES systems, the RAD platform may suffice in the short run, but it’s a temporary fix. The system is still legacy; you are still tying it together with disparate applications; and most importantly, you can’t upgrade, without significant cost and effort, the core MES functionality.
If you have a newer MES, a low code solution can also work, for example, providing additional worker guidance; adding an IoT platform for ‘edge’ data capture and analytics, or enriching the user experience with UI and UX applications. But again you have the challenge of maintaining two separate systems, two development environments, and two paths of upgrade and maintenance.
Another area where most RAD platforms are lacking is in the cohesive view of a manufacturing plant, with the infrastructure to serve both currents as well as future needs of the organization. A modern, extensible MES offers a platform with a comprehensive suite of fully integrated modules, designed specifically for the challenges of a manufacturing environment. It can integrate distributed devices, applications and infrastructure systems and add the context required from disparate data sources to enable advanced analysis and deeper insights that will provide a competitive edge in terms of efficiency, agility, quality and strategic decision making. Its broader view of ‘manufacturing’ allows it to not only focus on the manufacturing process, but outlying entities such as dispatching, quality management, production management and integration—areas where low/no-code may touch, but never replicate fully at this point.
Having a “no-code” platform may be somewhat misleading. Having little or no coding is an advantage but does not mean that systems can be implemented without a detailed understanding of processes, problem-solving skills and a logical approach. For large, complex applications, levels of abstraction simply cannot and do not currently go far enough. An efficient and effective manufacturing solution, especially in high-tech industries, requires a detailed understanding of functions at a coding level.
MES solutions designed for the future optimize efficiencies and provide a pathway to Industry 4.0 manufacturing, with extreme levels of automation and deep understanding of big data from across the shop floor, through to multiple manufacturing sites and into the wider supply chain. RAD platforms are fundamentally constrained with reduced code optimization, many with canned libraries and code, and potential tie-in to a single vendor.
The best of both worlds
The latest MES solutions incorporate a low code/no-code platform, providing users with the core manufacturing functionality needed while offering the flexibility to quickly generate purpose-built complementary apps, providing custom functionality not covered by the core MES. This low code platform within the MES can enable manufacturers to respond to cultural changes and quickly adapt to their unique challenges.
RAD platforms are a relatively new technology for manufacturing. The promises that they hold—faster development of worker-specific apps; lessened dependence on external developers; and the flexibility of adding functionality as required—all fit the needs of manufacturing who must quickly respond to market changes and customer needs.
However, there are some shortcomings: lack of specific domain/industry knowledge; single vendor lock-in; lack of a cohesive plant model and risks around security and scalability (especially when considering multi-site deployments).
Nevertheless, MES in the future will need to have low code as this is a primary means of extending the solution beyond the pre-built scenarios or the predicted levels of flexibility. By incorporating a low code feature in the MES, it enables a simple way to create speciality, point solutions that fall outside of the domain of the MES. It allows coding to be performed by citizen developers instead of pure IT resources, saving time and embedding ‘best practices into the system.
Low code generated apps will not replace the comprehensive application suite found within the MES. Rather, they will play a supporting role to extend the value of MES and provide a straightforward way of semi-customizing the functionality, allowing the MES to evolve to stay current with the businesses’ needs.