Noël, a passionate Data Scientist with a background in industrial engineering and a specialization in data-intensive industries, shares his experiences and insights. With over a year of experience at Nobleo and a focus on applying quantitative analysis methods, Noël offers his perspective on the role of a Data Scientist within Nobleo Manufacturing.
Why did you choose Nobleo Manufacturing?
“I chose the Data Scientist position at Nobleo because of my interest in Data Science, combined with my desire to stay engaged with the business context. I didn’t just want to develop technical skills but also apply practical insights to create value.”
“At Nobleo, I find an ideal balance between technical challenges, client interaction, and creative freedom. Additionally, I appreciate the flexibility and collaboration within our team. We closely collaborate on various projects, leveraging and valuing each other’s expertise.”
What project are you currently working on?
“Currently, I’m involved in a long-term project where my role involves applying quantitative analysis methods to generate practical insights for our clients. One outcome is the development of a planning tool for a client, where we combine complex factory simulations with genetic algorithms to generate optimal production schedules.”
“To accomplish this, I work closely with both the Nobleo project team and the client. One of the challenging aspects of my role is translating complex analytical concepts into understandable language for the client.”
What is the approach to such a project?
“We always try to approach a project in a structured manner, but the project scope needs continuous management. In the project I’m currently working on, besides developing the tool, many other client questions arise: for example, I’m involved in determining the theoretical end state of a factory. The client aims to increase throughput in the factory and wants to know: what is our maximum capacity? To estimate this, we used the simulation for the planning tool. We looked at: what is the bottleneck within the factory? How can we improve it? Is it realistic? Ideally, you don’t need to test in the actual factory. You test it in the simulation and then see the effect. This way, we’ve identified quite a few improvements. It’s important in a project to continue working from the problem/client question. We focus a lot on the client and what solution fits them best. This makes projects quite dynamic.”
How do you then translate the concepts and findings to non-technical stakeholders?
“It’s often useful to structure or visualize data. Creating a schematic representation, for example, of the architecture of the planning tool helps the client understand how the tool works. You can clearly show where the information comes from, what we do with it, and how everything is connected. I also enjoy visualizing data, to create simple insights from complex data, for example, through a dashboard.”
“It’s also interesting to focus on the usability of a tool. So how is someone going to use such a tool? It must be intuitive and reliable. You have to consider quite a few aspects.”
What do you typically do daily at Nobleo as a Data Scientist?
“On an average day, I spend part of my time developing a tool. There are always improvements and/or adjustments to be made. Additionally, I’m mostly engaged in conducting experiments. It concerns the ‘final state’ of a factory. New insights that the client asks for need to be processed. You translate those into a set of experiments and conduct those experiments with the simulation. We also have weekly meetings with the team and with the client. When developing the tool, I collaborate a lot with colleagues, so there’s a lot of brainstorming.”
What advice would you give to potential candidates?
“Data Scientist indeed sounds quite technical and analytical, and you do need to be that. But I think the business side is also very important. Because it all seems very easy from behind your computer, but there’s obviously a lot more to it. All the problems you’re presented with at university all have a perfect answer. But in reality, that’s not the case; it’s always a bit of weighing things up. What fits best? And there’s not always one very clear answer. This means it’s not enough to simply interpret the results of a data analysis; you also need to translate them into reality. Therefore, it’s essential to understand and appreciate the business context.”
Noël represents the profile of a dedicated and versatile Data Scientist at Nobleo Manufacturing. His expertise, enthusiasm, and commitment to continuous growth make him a valuable addition to the Nobleo team and contribute to Nobleo’s success in delivering high-quality solutions to clients.