Client background & business context

  • Client: Global mechanical and plant engineering firm 
  • Project: Error prevention: Feature to error correlation – Continuous data monitoring for error detection and prevention; tactical algorithm deployment at our client’s facility 


In a manufacturing ecosystem where perfect functionalities, top-notch features and highly qualitative aesthetic delivery are essential, our client’s challenge was to come up with an effective solution to improving the overall quality of its paint shop process. The goal was to rapidly reduce the number of car body parts with painting faults and the assumption on the matter was that an analytics algorithm could detect if a specific step in the painting process will lead to an error for a particular car body part. 

Our approach

The vision of deploying a tactical algorithm for error detection quickly came about with the help of our dedicated team of data scientists and software developers. Since every successful initiative relies on full comprehension of the context and an honest collaboration, we made sure from the very beginning that we nurtured a strong and consistent communication of goals, needs, milestone metrics, iterations.  

Hence, our team was involved in every software development phase, starting with prototyping the algorithms, evaluating them on test and validation datasets and, finally, deploying them into our client’s infrastructure. The communication bridge allowed our team to lead some of the most important decisions in terms of algorithms development, implementation and ongoing tuning, according to the established milestones.  

The feature to error correlation implied a continuous monitoring of the qualitative data generated during the painting process. Based on the data provided by the client’s monitoring staff, our algorithms would analyze the data, detect if a certain combination of car body features or settings in the paint shop process leads to errors and notify the user about a possible error cause.   

At the same time, we could develop the algorithms to also detect periods with high error concentrationhaving the data provided by the monitoring staff, the algorithm would analyze and detect which time frame was prone to having a higher error rate. 

Our commitment to a highly qualitative customer experience translated into: 

  • Continuous researching of state-of-the-art analytical algorithms and solutions which could be integrated in the client’s product 
  • Proposing custom-built solutions for candidate algorithms 
  • Continuous communication in order to evaluate if our solution met our client’s goals 


Data storage: 

  • MongoDB 

Development tools: 

  • Python 
  • Pandas 
  • NumPy 
  • Matplotlib, Seaborn 


The impact of our solutions was rapidly perceived: we have deployed tactical algorithms that allow our client to analyze quality and error data for more than 20.000 car bodies, in time intervals ranging from less than 2 minutes to 10 seconds. The algorithms are executed on a weekly basis and, depending on the data generated in a specific week, the users are notified if a certain feature leads to error or which periods might have higher error rates.  

In the quest to improve the quality of our client’s painting process, we offered tools that quickly identify problematic circumstances and delivered a more efficient approach to detecting and, subsequently, preventing operational errors.  

Related case studies

Project Reporting Process Digitalisation

Cloud application to support the digitalisation and automa ...

Cloud-First Strategy: Microsoft Exchange to Office 365 Mig ...

On-premises Microsoft Exchange infrastructure migration to ...

API first integration for enhanced availability and reserv ...

Receiving notifications about any possible breaks in the ...


Mies-van-der-Rohe-Strasse 6, c/o Design Offices
80807 München, Germany
E-mail: hello (at)


Constanta 12, Platinia
400158 Cluj-Napoca, Romania
E-mail: hello (at)


Cetatii Square 1, Oradea Plaza
410520 Oradea, Romania
E-mail: hello (at)