Decision Support for Beekeepers challenge


The decision support for beekeepers challenge will focus on honey bees. What are the relevant relationships with environmental factors, and how can they be translated into a tool that helps beekeepers to optimise choosing hives locations? This also may be relevant for agriculture, and crop growers who depend on pollinators.

If the suitability is low, the question becomes what can be done about it. For example: different mowing strategies by municipalities, tree planting strategies to optimise gestation throughout the season, temporarily moving hives or, if necessary, supplying temporarily provide additional feeding options to bridge critical periods.

We also want to look at the question of whether honeybee population dynamics could imply things on the quality of the local environment. This is also important for wild bee species.

Besides environmental factors, interesting dimensions to take into account are:

  • time factor: bees need feed from spring to autumn. What is the availability of feed throughout the season?
  • the weather: depending on the weather, plants and trees bloom or not (eg too dry as the heather in The Netherlands this year), sooner or later (trends show that spring starts earlier every year in the Netherlands), etc.

A valuable result of the hackathon might be that a beekeeper would be able to click on a location, and then sees a certain score for suitability. For example:

  • a low score, with explanatory parameters such as low plant gestation in autumn. The user then knows for a healthy honey bee population that he either needs to improving gestation in the fall, move hives or add other feeding options.
  • a high score, with an underlying explanation. Maybe even an indication of how many hives the region can hold.

It would also be really cool if the user can also add relevant insights and experiences in a particular location, give feedback on the score, help detail the vegetation, or even allow access to data on how much honey he yielded and the number of colonies he/she keeps. This type of data help to validate and further improve the decision support model.

📈 Data inventory 🗂
About the Big Data for Bees Hackathon
Who is Who at the Big Data for Bees Hackathon

I think it would be a good idea to not only optimize ‘choosing hive locations’, but also optimize the number of hives which can realistically be managed in a sustainable way in a certain location.This might help the individual beekeeper, the beekeeper associations and councils to focus on the problem of sustainable beekeeping, as well as leading beekeepers and other parties towards the question what can be done in their area to make it more suitable for honeybees. It might also mean that beekeepers (as well as councils) have to decide what is fair in allocating places for hives, working together with people who know about biodiversity.
It would also be good if the specific threats beekeeping poses to wildlife in that particular area are listed, so the se can be addressed appropriately.


Hello, how is it going? Ready to pitch to the jury? :slight_smile:


Ha. Almost. Deep work going on here.


Committing and prepping for pitch, face2face on Github and on Slack


Team photo


Our hack can be found at, the data sources are Food4Bees indexkaart and Apiary Map. Beespot code shared on Github here.


Cool, thanks! That looks like a tight operation!


Second prize!! The Food4Bees foundation (gentleman 2nd from the right on the top row) wants to partner up to take this idea further! Thank you all!


Hi all! just wanted to share with you the work done by @mvmacke during the hackathon, it kinda got lost in the Slovakian presentation on the IoT solution, but he did some proper analysis on the difference between desired pollination and actual pollination, for different areas (riverclay, sandy soils, etc) in the Netherlands! /c @mato74


that looks so pretty. This kind of analysis is possibly very interesting to include in the farmers’ dashboard for monitoring pollinator hospitality
Therefor also relevant to @Inge, @R0bKnapen, @Turan @Turan_Bulmus and @IreneGM and Herbee people (@lucas? @JacomijnPluimers)


Perhaps @mvmacke is willing to share here a little explanation about the type of analysis, the data used, and the tooling?


Tjeez, I’m such a newbee. Ha, ha, I posted on the Data Inventory:

Hi, just shortly about the hack we made for “To Bee or not to Bee”.
We used a different visualisation than a map. We used treemap which is a visualisation of quantity. The best software for this purpose is the Treemap software from the University of Maryland. It is Java so it is cross-platform. Really cool. What makes their software so nice is two things:

  1. at the highest level, you can see the lowest level elements
  2. it has fantastic filter possibilities which can be shown while you create the filter
    The datamodel behind is really simple: just rows and columns than one empty column after which you can create your hierarchy. Extremely cool

We used it to show the data that were in the labur ?? model from the WUR (I see they are already closed…). Very cool data since it shows wild bees per type of landscape. We faked the time aspact by introducing some random numbers and weights. I hope you don’t mind…

In general I find that there lots of good data on bees and others but it could be much better disclosed. I would love to work on that to make nature talk a bit clearer and louder… :wink:

So can I help anyone to make nature talk?? I know this is a very general remark but I can’t get it more specific!!!


Ok, geen Labur model maar LARCH. Close, toch?..


Respect!!! But with Stefan on your time, it is easy to win!! :wink: