Hi Valentin,

Adding more to Rayan’s point and also elaborating a bit. As a feasibility study you could check over internet if the problems are solved by others. Are there any clue/logic/open-source implementation of them? They could be your exact problems or similar to them. The respective developer can understand the ways. This will help you to identify the blockers if any. Because blockers are the ones which shoots you away out of your estimation. Once the logics are understood break the problem statements into various small demonstrable outputs and estimate them. For any Output there is always an easy part, average part, hard part and harder part. Easy or certain part which you can estimate correctly, because it done earlier, but add few buffers time to hard + harder part.

So, if you don’t have any expert to suggest & estimate, above is way you could approach.

As Rayan mentioned you should always balance between your good/exact estimated time-line and bagging the project. Sometime you need to estimate tight (effective/billable hours) in terms of providing cost but could stretch the calendar deadline through project management negotiations.

Thanks,
Sandeep