Prototyping is essential in the development of software for intralogistics, allowing teams to test ideas, refine solutions, and enhance user experience before full implementation. In this article from our Product Management series, Georg Held, Product Manager at arculus, shares how we use prototyping to drive innovation, optimise software performance, and improve efficiency.
Prototyping in Intralogistics
Intralogistics systems, supporting warehousing and production require a well-integrated hardware-software solution. While many monolithic systems perform well, they often hinder innovation. Exploring, testing, and refining new ideas before implementation is challenging. To address this, arculus leverages prototyping to gain insights before adopting new solutions. This allows us to collect early feedback, minimise risk, and improve efficiency to maximise product potential.
Let’s explore how arculus uses prototyping to innovate in planning, installing and managing a fleet of Autonomous Mobile Robots (AMR).
What is Hardware-Software Prototyping?
Today's prototyping culture goes back to Eli Whitney's promotion of "interchangeable parts”. In the late 1800s he advocated standardising identical components for easy replacement. This enabled mass production, simplified repairs, and improved integration and testing of new parts. It also allowed cost-effective iterations to enhance products in development, or for already existing ones, forming the foundation for modern prototyping.
Centuries later in the 1990s, the term hardware-software co-design emerged, as embedded systems gained recognition. Today arculus builds on this concept with a strong prototyping culture. Hardware-software prototyping as the name suggests, combines both hardware and software, allowing users to test products before full production. This approach enables effective testing and fosters innovation, efficiency, and clarity. There are many ways to engage users:
- Physical models, even non-functional ones, are used to represent form and design;
- Mock-ups that demonstrate the visual design of a concept;
- Simulators that replicate system behaviours;
- Clickable prototypes that allow users to interact with the system.
Numerous benefits such as the early identification of user experience problems, justify the time and effort required to create and test prototypes. Feedback from these tests is critical for future product development iterations, and establishing a learning culture through prototype experimentation is essential. As discussed in the previous blog post, prototypes are a key part of product discovery at arculus.
The Strength of Prototyping at arculus
The high complexity of intralogistics systems requires a strong prototyping culture. At arculus we rely on it not only for hardware but also to guide product development. By incorporating prototypes early we gain valuable insights from the start, enabling continuous evaluation and necessary adjustments throughout our discovery process. This agility ensures we integrate feedback early on, optimising resources while refining user experience. Crucially, recognising when to stop prototyping after validating a hypothesis ensures efficiency and keeps product delivery on track.
From Wireframes to Clickable Prototypes
After agreeing on the user journey we immediately moved to wireframes of the desired functionality, adding brief descriptions for context. This created a collection of screens covering key parts of the user interface, while revealing gaps to address. Using a collaborative tool like Figma, we efficiently captured extensive feedback in real time, enabling continuous iteration.
To better reflect the user experience, we compiled a clickable prototype by adding interactivity to the wireframes. This helped overcome challenges from our legacy software toolchain, as the previous solution lingered in people's minds. The visual, and more importantly the interactive approach, encouraged users to explore new solutions, helping them quickly adapt to move beyond past systems and actively engage. In the end, the validated prototype answered many of the questions that our development teams had, as they built the real application.
Understanding System Performance
Later on, and as an important part of the user journey for the arculus Fleet Manager, we simulated a fleet of mobile robots, to see the heart of the software in action without connecting to real vehicles. This framework helped us explore possible transport scenarios for the first version of the application. Using a challenging transport matrix, we configured a large-scale simulation scenario—another way of prototyping that allowed us to understand performance on all levels of the software at an early stage in the product discovery and development phase.
In this phase we validated the required infrastructure, architecture, and algorithms for the arculus Fleet Manager. We set up dozens of driveways, deployed 100 mock-up vehicles, and established 1000 handover stations. The simulation ran smoothly for hours, and the new software toolchain achieved the goal of having a stable foundation that optimised material flow in intralogistics for our customers.
The prototype proved that the core components of traffic management were suitable for large-scale deployment. However, for the MVP (Minimum Viable Product) we opted for a more robust, purpose-built simulator developed from scratch.
In one of our future blog posts, we will provide insights into the simulation capabilities of the arculus Fleet Manager.
Connecting the Dots: From Prototype to Physical Movement
The next step was connecting to real vehicles. Since the software on the robots already had defined interfaces and communication protocols, we could quickly perform the first driving actions. Naturally we recognised the need for a more robust interface between the arculus Fleet Manager and the vehicles for real projects, but taking a step-by-step approach reduced the complexity of software-hardware interaction early on.
Instead of defining and implementing the entire interface at once, we iteratively developed the most important functionalities for validation. One goal was to confirm that the new arculus Fleet Manager, with its high user experience standards, could guarantee safety and accuracy in real-world scenarios. For instance, through prototyping we tested the hypothesis that users wouldn’t need to enter error-prone coordinates, but could rely on a visual interface for the alignment of all layout objects. Benchmarking vehicle positions between the layout and reality took several iterations, but each step confirmed we were on the right track.
The prototypical approach also helped us gain the trust of critical users, especially the field engineering teams responsible for commissioning projects at customer sites. Beyond validating the initial goals, prototyping also revealed other crucial improvements to the user journey. To capture those during typical testing sessions, we created systematic scripts with expected results, leading to clear actions for related teams. And half a day later the backlog already reflected new items for upcoming sprints. Rigorous prioritisation enabled teams to stay focused, and subsequent test sessions confirmed that development was moving in the right direction.
In an upcoming blog post, we will share more insights on how the arculus Fleet Manager’s user interface supports rapid project deployment.
Our Recommendation
There is no one-size-fits-all approach to prototyping. So here is our recommendation for getting the most out of it:
- Select and agree on the component or feature to prototype. Often prototypes are not for the most technically complex topics, but for those requiring the right user experience. Focus on the hypothesis and goals you want to achieve and strictly adhere to them to complete concepts quickly.
- Prioritise between low-fidelity and high-fidelity. Although prototypes support more efficient development, consider how much effort and complexity to invest. Low-fidelity prototypes, even on paper are often enough in the early stages of product development. Remember: the priority is to validate the concept, not how you build the prototype.
- Discuss and choose the right prototyping tools. The most suitable tool depends on the type of feedback you need. Over the years we've learned not to limit ourselves to a few tools, but to embrace cutting-edge solutions, and to stay open to trying new ones. Also, don't feel obligated to reuse the same prototype for subsequent development stages.
Finally with the advent of generative Artificial Intelligence (AI), it has never been easier to start prototyping. AI not only aids in creating prototypes, but also in quickly analysing and evaluating feedback. As this technology evolves, we continue to explore innovative ways to prototype, accelerating our product development in the future.
