INDUSTRIAL DIGITALIZATION IN SMES: how to apply Industry 4.0 in a practical way | DataTalks #4
May 29, 2026
RELIABLE AND SIMPLE DATA IN INDUSTRIAL PLANTS:
DECISIONS BASED ON REAL DATA | DATATALKS #5
Videocast DataTalks | Episode 5 with Iván Carbia (Operations Director at Metaldeza)
🔍 In this episode of DataTalks, we chat with Iván Carbia,Operations Director at Metaldeza (Metalúrgica del Deza), an industrial engineer with more than 15 years of experience leading operations in sectors such as automotive, aerospace, and manufacturing.
Throughout his career, Iván has worked in different countries (including Northern Ireland and England) managing multicultural teams and highly regulated projects, which has allowed him to develop a very solid vision of data, processes, corporate culture, and decision-making in complex industrial environments.
In this episode, he shares a very clear and practical vision of how to work with reliable and simple data in an industrial plant without falling into complexity, applying principles of Lean Manufacturing, Six Sigma, and continuous improvement, and always focusing on ensuring that the data is reliable, understandable, and useful for people.
🧠 Why is this relevant for Operations Directors and Lean Managers?
During the conversation, many common concerns in industrial operations arise:
- Why do we have so much data but still make bad decisions?
- How can we know if data is truly reliable before acting on it?
- What role does culture play in data-driven decision-making?
Iván answers from his extensive experience in operations, lean manufacturing, and continuous improvement:
- More than 15 years leading industrial operations
- International experience in sectors such as automotive, aerospace, and manufacturing
- Practical application of Lean Manufacturing and Six Sigma
- Clear focus on simplicity, reliability, and usefulness of data
🔍 Episode highlights - Key ideas
1️⃣ Reliable data in industry: the basis for any decision
Iván explains that many industrial decisions are made based on data that is not actually reliable.
If the data is collected incorrectly, measured incorrectly, or interpreted without understanding the process, the subsequent analysis loses all its value.
In several examples from his industrial experience, he shows how the quality of data in industry is more important than the quantity of information available.
2️⃣ When the problem is not the process… but how it is measured
One of the most illustrative cases in the episode shows how 25 years of industrial data were based on incorrect measurements.
The problem was not in the production process, but in how the data was collected:
inconsistent manual measurements, errors in recording values, and procedures that no one had questioned for years.
When the measurement system was reviewed and the data capture process simplified, the results changed completely.
3️⃣ Data-driven Lean Manufacturing: going to the plant to understand the process
Iván insists on the importance of going to the plant and observing directly how the data is generated.
Concepts such as Gemba Walk or Go to See are not just Lean terminology: they allow you to detect errors in data collection, informal practices, or deviations between the procedure and reality.
Only by understanding the actual process is it possible to build a data-driven continuous improvement system.
4️⃣ Action-oriented industrial monitoring
In many industrial environments, reports and dashboards are generated that do not change anything in the operation.
Iván argues that industrial monitoring should be action-oriented, not just for viewing indicators.
The data should help detect problems, prioritize actions, and improve the production process.
When data does not lead to decisions or changes, it loses its operational value.
5️⃣ Simplicity in industrial digitization: the KISS principle
One of the clearest messages of the episode is to apply the Keep It Simple, Stupid principle.
Data systems should be:
- easy to understand
- easy to collect
- easy to use on the plant floor
When the system becomes too complex, teams stop using it and the data loses reliability.
Simplicity thus becomes a competitive advantage in industrial environments.
Final tip
A good data system is one that is understood, trusted, and used. When data is reliable, understandable, and used on the shop floor, continuous improvement flows naturally.
📺More episodes of DataTalks
At DataTalks, we talk to industrial, operations, and innovation directors about: Industrial digitization, Industry 4.0 in the plant, practical use of data, and, above all, real experiences without unnecessary theory.
Do you want to get the most out of your plant data? 
CTO & TECHNICAL DIRECTOR
Expert in industrial monitoring and data analytics.
We tell you how to improve decision-making and production efficiency in your plant, without wasting time generating reports. Your plant at a glance!
Subscribe to our Newsletter
