Agricultural production optimization

Vikima Seed is the leading producer of vegetable seeds in Europe and USA. The company produces vegetable seeds of the highest possible physical and genetic quality to fulfil the demands of the customers worldwide.

In Vikima production, there are three cleaning lines that remove weed and plant material from the harvested seeds and ensure purity standard of above 99.8%. The line is an array of different cleaning machines that progressively purify the seeds to production quality.

Due to the high variance of seed types, traditional equipment for flow measuring cannot be used on the lines. The lines have been running without any throughput measurement other than just counting boxes of cleaned seeds, which was usually “one per hour”. This is not sufficient production data at all.

The customer was in dire straits, positively

The customer was facing a + 40% order increase for the 2017 season, and had already been producing 24/7 in the previous season. Installing new cleaning equipment was expensive, took a long time to erect and would have only inherited the “bad habits” of the existing lines.

Then, Vikima Seed hired consultants from Jens Jochumsen Advisors to optimise the existing lines to cope with the extra 40% production. Even with the existing scarce production data, it was obvious that throughput was very fluctuating over the time.

Changeover time was optimised through a Lean and SMED project, which increased possible production time by 20-25%. The remaining +20% were to be pursued in optimising the production throughput.

The consultants needed facts - actual production data.

You cannot optimise if you do not know the current status.

The consultants developed a simple flow meter to be installed at the end of the cleaning line before filling into the 2 m3 boxes. The flow meter is a buffer container on top of a rotary valve. Level sensors in the buffer container control the rotation start and the speed of the rotary valve while they also ensure full chambers under rotation. This meant that they only needed to count number per minute to get an accurate throughput measurement, knowing the volume of each rotary valve chamber.

Factbird was perfect!

The cleaning lines are not equipped with modern Scada/PLC for data handling so it is a huge work to add the flowmeter in the production control system. Instead of upgrading the legacy system, the consultants chose to use Factbird to measure the chamber’s rotation. It was extremely easy to implement because the Factbird is a completely independent system and works outside the existing control system.

Auto data handling and visualisation on Blackbird application

The Blackbird application handles the data from the Factbird and shows critical insights, which is the key success factor for the project.

After a very short time of setting up on the Blackbird web application, all the production data they needed is right on their hands. The throughput is now shown as live data presented in simple and useful charts. It is even possible to enter stop causes in less than 20 seconds which gives vast data source for analysis.

Impact on the operators

The first major result of the flow meter and the Blackbird application is the impact on the operators. The customer now has direct and detailed information of how the line is running. The customer can now optimise and balance the line to a degree that was impossible before since the performance of the cleaning lines was previously evaluated based on feelings and semi facts.

Being able to know stop times and the causes are also now bringing positive result. The analytical tools of Blackbird quickly give an overview of which actions to pursue.

There is however a trip wire. The hard part is to make the operators remember entering stop causes. This requires constant follow-up until the culture is changed.

Made it !

One month into the season, the customer together with the consultants successfully improved the production capacity as required: extra 40%.

Now that the customer is getting accurate production data from the Factbird, there is still a lot to do…. The production data shows a lot of possibilities for further improvement.