Data interest for my Bachelor thesis

General
1

Hello.

I am currently writing my Bachelor’s thesis on “Return on Investment for the Development of Open Source Science Hardware: The OpenFlexure Microscope as a Cost-Effective Alternative to Standard Microscopes and Manufacturing Optimization Opportunities.”

To calculate the Return on Investment (ROI) of the OpenFlexure Microscope, I need the following data:
1. The time-dependent number of downloads of the OpenFlexure design files.
2. The estimated percentage of downloads that result in a completed product (e.g., assembled microscopes).
3. Marginal costs for digital manufacturing (e.g., 3D printing or CNC machining) of the OpenFlexure Microscope.

I would greatly appreciate any guidance on where I might find this data or if anyone in the community has access to relevant information and would be willing to share it.

Thank you for your support and insights!

2

@Silas-bue it is clear that if you have those three things, then it would be straightforward to understand a monetary return on investment. However in the age of crawlers and bots, item 1 tells you almost nothing, the numbers can be huge. Even in this Forum, more pages are served to bots (search engines etc) than to people. Item 2 is unknowable for an open project like this - successful builds do not even show up as requests for help here. Item 3 you can easily work out yourself from the instructions, preferably building one yourself to understand any time costs.

This Forum shows activity that is visible and public. Can you find any studies that have considered the proportion of activity that comes to any Forum in general - for hobbies, software or anything else? If there are some clear multipliers then you might get somewhere.

3

If you haven’t already, I would start by downloading the files, buying a kit of parts, and making your own OpenFlexure microscope. That will give you a greater insight into the aspects involved than you can get from the online resources (great though they are).

Some initial thoughts on your questions:

As @WilliamW notes, the number of ‘downloads’ doesn’t really relate to much: you could have an individual who downloads, and then builds a microscope from it, or a company which downloads it once, and makes a thousand. And lots of downloads which don’t result in builds because they were bots, or were people who haven’t got round to it yet.

While it is not strictly essential, many builds of the microscope will likely include a Sangaboard to control the motors. I am not sure whether the data on this is commercially sensitive ( @filip.ayazi ?), but it may be possible to obtain the sales numbers for this, and with suitable caveats that may act as a proxy for build numbers (and the trends over time may also be informative).

Manufacturing costs are easier to quantify (side note: CNC machining isn’t an option for this, it can only be FDM 3D printed). Many newcomers to 3D printing look at the amount of filament used, and erroneously base their costs solely on that. However this will be a small fraction of the overall cost, as there are many non-printed parts you need to include, and there is also the depreciation on the 3D printer and the everything associated with the workshop environment used, plus of course the costs and time taken for assembly, testing, certification, sales, marketing and delivery.

I guess an underlying question is clarifying the title of your thesis, “Return on Investment…” ie who you are looking at to gain the RoI? It is the end-user who is benefitting from obtaining a lower-cost alternative, or a manufacturer who can undercut the competition? One of the drivers behind the development of OFM is reaching markets that could otherwise not afford such devices at all - so in areas where it is not just an alternative to ‘standard microscopes’, but an alternative to not having one at all.

Good luck with your thesis. I for one will really look forward to reading it.

2 Likes
4

Hi @Silas-bue,

Following on from what @WilliamW said. One of the things we have concentrated on most with OpenFlexure is being truly open, which means not only sharing all information, but also giving users as much freedom as possible. Part of that freedom is freedom from us tracking them.

This creates a bit of an unfortunate situation, because while we believe this freedom makes the project far more impactful. Itl becomes significantly more difficult to measure that impact! Many pieces of modern hardware and software regularly “phone home” telling developers what features are being used and when. This has benefits to the development, but at the cost of user’s freedom and privacy.

At some point we may add an optional extension that is 100% opt in (because you need to install it yourself) that sends some anonymised feedback to developers. But this still wont tell us about numbers, as most wont opt in.

So how do we track impact? We have our Where Are You thread on this forum, which gives us an idea of graphical extent. We can also search and look around both the internet, and around Google Scholar.

From this we can see 100s of people are on the forum, microscopes have been built in all continents, in almost 60 countries (and this is only the countries of the people who have told us, so it ill be more).

From all of this some numbers can be estimated probably, but it will take some creativity! And will have uncertainties!

1 Like
5

Thanks for sharing your approach - it is indeed a good idea to go through the community members. I use J.M. Pearce’s methodology, which is specifically designed for open source hardware projects like this one. While it does not allow you to determine an exact value, it does provide a solid basis for making a reliable estimate.

With the addition of the other two methods Pearce mentioned, I think I can get an even better comparison. I also agree with your point about marginal costs. I plan to determine these based on the work that has already been done to set up production.

Thanks again for your thoughts - they are much appreciated.

6

Thank you for your detailed response and the thoughtful points you’ve raised. I’m happy to say that I have already built an OpenFlexure microscope myself and am actively researching with it. This hands-on experience has been invaluable in understanding the practical aspects of its design and operation.

Your suggestion to look at Sangaboard sales as a potential proxy for build numbers is exactly the direction I could explore further. As the Sangaboard is often essential to the operation of the microscope, its use could indeed provide a clearer picture of adoption trends and active setups. Do you have any idea where I could find such information?

Regarding manufacturing costs, I appreciate your reminder that filament usage is only one component of the overall cost. Factoring in non-printed parts, equipment depreciation, assembly and other operating costs is something I will carefully consider.

As for the focus of my thesis, it is centred on the perspective of the end user, particularly those who would not otherwise be able to afford such equipment. The aim is to show how the OpenFlexure microscope can serve as a cost-effective solution in underserved markets, bringing scientific tools to communities that might otherwise go without.

Thank you again for your encouragement and for sharing these insights. They are incredibly valuable to my research.

7

Thank you for your email and your valuable suggestions. The aspect of freedom is indeed a fascinating point that I hadn’t considered before. It has given me a new perspective that I plan to incorporate into my work.

I also now have a better understanding of why some of the data is not publicly available and will do my best to use the available information in a clear and understandable way. I also plan to do further research using the WhereAreYouThreads section of the forum and hope to generate some relevant figures through targeted internet research. With a bit of creativity I will try to come up with a reasonable estimate.

Thanks for your helpful hints and support!

1 Like
8

I think the best person to approach is Filip Ayazi @filip.ayazi (who will hopefully spot I have tagged him here. If not, @WilliamW or @j.stirling may be able to put you in touch)

I would be quite interested in this data too, so if you are able to share it with me once you have it, that would be appreciated.

2 Likes
9

I would strongly caution against Joshua Pearce’s costing strategy. It has a tendency to massively underestimate costs as many of his costings ignore not only the fractional cost of the machinery and space needed for manufacturing, but also assumes all time is free.

By such a metric this fence I made is free:

2024-01-27 16.40.14
2024-01-27 16.40.141920×1440 393 KB

We chopped the wood out of our own trees, we already had the tools for cutting the wood, stripping the twigs. However, if I tried to build a business selling these fences on the assumption they cost nothing to make, I will be very disappointed when I realise that I have to pay the people that build it, and buy enough land to grow coppices for wood, manage the coppices, and maintain my tools.

I do think that OpenFlexure does result in a very good value instrument, even when you factor in time taken. I think it even represents very good value at a global level even when you factor in the time that has been taken on design, considering that we estimate that thousands have been built.

I think part of the “value” benefit is in creating an instrument that can be customised to a new uses easier. Creating an instrument that can be maintained for longer. Creating an instrument that can be adapted to different environments. These values are harder to capture by simple formulas. The probably take a bit more of a qualitative approach, which can be enhanced up by estimated numbers.

It would be an interesting number. But it would likely underestimate the number significantly as I know many people across the world use the Arduino work around. Personally I have built 10s of microscopes none which have a new Sangaboard from Filip. Probably in a few years this may begin to converge as Sangaboards become more available, or it may diverge as people use Sangaboards for other non-openflexure projects.

2 Likes
10

Thank you for your input. This is definitely something worth considering, and I will take a closer look at it moving forward.

11

Fantastic conversation. I enjoyed reading it. My 2 cents: consider the cost of the printer and electricity as a variable too.

1 Like