Logo de Icube SA à Bulle



Increase the profitability of your production, reduce waste, use less chemicals, perform preventive or predictive maintenance. These are all challenges facing production managers today.

But in order to meet all these challenges, there is an absolute prerequisite: to have the data from your machines fed back into your information system thanks, for example, to the IoT (Internet of Things).
Most recent machines are equipped with tools for sending data.

The challenge in this case will be to consolidate the data from a heterogeneous fleet of machines and then to know how to interpret all this data through common indicators (KPI).

For machines not originally equipped with such tools, several questions arise:

– For what purpose do you want to connect these machines and what is the expected value?
– How do you analyse and value these data?
– How do you maximize the value of this project?
– Do I need to be accompanied to carry out this project?

In this article, we will try to accompany you to best initiate your IoT project.

What should you expect from IoT?

The main interest is the collection of machine data, at all times and on a continuous basis.
Today, it is possible to know exactly how a line is performing, easily identify breakdowns and the steps that slow down production.
For example, it can be noticed that a control station often gives a bad part when in reality the part is correct. So by making a small adjustment to the control tolerance, a positive effect on production can be measured by comparing the feedback data.

It is also possible to follow the functioning of the different elements of production and monitor their use..
In the event of a stoppage or breakdown, the technician can study the case remotely and perhaps even solve it without having to travel. For example, by having access to the machine from his workstation, the technician will be able to see directly that a probe is probably no longer working. He will then be able to temporarily disable the remote control, or go on site to change the part, knowing in advance what equipment to put in his vehicle.

By reducing round trips, identifying problems more quickly and remotely controlling production, downtime is greatly reduced.

On the other hand, the data sent back from the machine are raw, so it is necessary to use a software that allows to visualize these data correctly in order to be able to read and process them efficiently.
For example, we would like to know the machine downtime during a specific time slot, in order to be able to see if there is a noticeable and regular difference between different shifts.

Or simply be able to view a yield curve to see if the latest changes made to the line have had an impact.
Feedback is an essential part of machine connectivity. Quickly it will allow you to target which actions to implement, and then it will allow you to track and monitor the changes made.

List and prioritize your needs

The first step in your project is to list and prioritize your needs. As far as possible, these should be expressed in a SMART (Specific, Measurable, Ambitious, Realistic and Time-bound) way.

Here are a few examples:

– Productivity increase of 20% by the end of 2020.
– Anticipate maintenance to reduce downtime by 12% within two years.
– Facilitate troubleshooting in order to reduce downtime by 5% by the end of 2020.
– Reduce your production waste by 10% at our main site before Q2 2021.

Prioritize your needs according to a single variable: the value you bring. We will explain in more detail how, using Agile methods, it is possible to set up a prioritized and valuable feature backlog.

For example, what penalizes performance the most is usually machine downtime. It would therefore be interesting to prioritize the functionalities that allow you to avoid these stoppages. Anticipating maintenance and facilitating troubleshooting will also help to reduce these stoppages.

The progress of your project: case study

Mr. Dupond, production manager at Usine SA, aims to increase production by 20% this year. He knows that his production line can deliver this production without any problem, but he has noticed that there is a lot of rejected parts and especially that the production line is stopped far too often.

Mr. Dupond understands that he needs to collect more production data to be able to analyze the problems affecting his production and contacts Icube SA to help him achieve his goal.

Together they prioritize the functionalities to be implemented:

– Installation of a nanocomputer to receive information and send it to the cloud.
– Data integration with an application for recording, visualizing and archiving production data.

Once the hardware is installed, it is time to let the line run for a while to get enough data.

Analyses of the data received show that the final control station has 15% errors and that line stoppages represent a 25% loss of production.

After discussion, it emerged that we will focus first on the control station because by simply making a few adjustments, the number of rejects should be able to be drastically reduced.

At the same time, we will continue to observe what is causing so many stoppages on the line.
The adjustment of the final control unit has already reduced the number of discharges from 15% to 5%. There is probably still room for improvement, but the effort required would be significant for a relatively small gain.

As regards line stoppages, a number of factors have been observed: it was found in particular that, for the slightest error, the operators call in the technician. In addition, there are long stoppages due to equipment replacement.

Two solutions are implemented, firstly better training of the operators on the line so that they can be more autonomous and secondly a list of the components to be changed as a priority so that they can anticipate their change, for example during a planned line stoppage.

After some time, these two solutions have reduced the production loss by 15%.
Mr. Dupond is very satisfied, because with the 10% less rejects and the 15% production gained by reducing downtime, he has even exceeded his target. In addition, he now knows what to focus on to continue increasing production.

Project methodology and implementation partner

AGILE methods are particularly effective in prioritizing tasks and maximizing value. By favouring short development and implementation cycles, they allow to keep a permanent contact on the progress of the project, by involving the stakeholders (integrator and customer) throughout the development of the solution. You decide what is most important to you, and can stop the project as soon as it has delivered enough value to you.

However, these methods require some theoretical knowledge and experience. This is the advantage of having a partner who masters them. In fact, he will be able to quickly set up the necessary structure and have control over the smooth running of the project while limiting your need for training.

Moreover, a partner will be better able to identify what is technically feasible at lower costs, and to be able to propose the best optimizations. Your partner also has certain technical tools: for example, if the machine has not been designed to feed back information or if some key information is missing, he will be able to access the machine program and related software.

Data collection means storage. Some partners will also offer you data hosting solutions as well as software for the history and analysis of this data, setting up alarms etc. To go further, they will probably also be able to integrate these data with your management applications such as ERP, MES, CRM etc. In order to facilitate the implementation of your project, prefer actors with a complete mastery of this type of problem, from the installation of the sensor to the indicator that the director will follow, including the storage, integration and use of this data.


If connecting a machine park can already bring a lot of extra value to your production today, it is important to consider that this will one day become a necessity. Indeed, the ever-increasing production constraints require high-performance tools and an increasingly optimised production line in order to achieve your profitability objectives in particular.

But it is also crucial not to skip any steps:

Determine and prioritize your needs using proven methods to maximize the value delivered by your project.
– Then set up the data collection and storage by having a partner with both IT skills and experience with machines accompany you.
Choose the right data visualization and analysis tool so you can easily determine what actions need to be taken and monitor their effectiveness.
Link your production data to your management applications to optimize your business in a cross-functional way.

Each company has its own history, each production has its constraints, its own trick, and this is what makes your product unique.

Wants to know more?