Questions

Here are some things we still don't understand.

Sorry if the list is rather long and looks like it's going to be a huge bunch of work, but it seems easier to just put all questions in one place and have you answer them little by little over time, whenever you can make it.

Here goes:

1. For scheduling your crops, it would be beneficial to subdivide your land into some sort of indivisible, smallest units such that at any given time, there is exactly one crop on such a unit lot (can you think of a better word?). For example, during some given week of the year you could have planted 5 unit lots of tomatoes, 10 of chards, etc. What I need to know is how many such "unit lots" would be reasonable to reckon with, given the fields you have.

To figure out their number and size, these lots should be small enough so that the program can assume that only one crop will be grown on each of them at any given time, yet big enough so that the smallest amount of any crop you will ever plant takes up a small integer number of unit lots (ideally, just one of them).

**EDIT Julian Nov 23** In fact, the unit lots don't have to be the same size. You could also allocate a certain area for tomatoes, another different area for beans, etc. You can also tell me the area in square meters, or meters of length, or whatever works for you.

2. Related to this, in the fields you have, can any crop grow anywhere? Since this is probably not the case (because, for example, beans need scaffolding but lettuces don't), you might want to start thinking of some kind of numbering scheme to distinguish the different unit lots, in order for the program to be able to incorporate restrictions on what can grow where.

3. I don't understand crop rotation a lot. Does it work like, if crop x is planted on lot y in week z, then starting from week z+z', at the latest, there must be a different crop on lot y? Or do you need something more sophisticated such as having three or four crops cycle in time on each unit lot? (I think that wouldn't be a problem to incorporate, either.)

4. Precedence constraints. Some tasks must be executed before others; for other pairs of tasks the order is immaterial; some tasks are mutually exclusive. Could you please fill in the following table? Put "before"/"after"/"excl" in row r, column c, if the task in row r must be executed before, resp. after, resp. is exclusive w.r.t. the one in column c; for example, weeding comes after tilling, and setting rows comes before setting up irrigation. If the order is immaterial, leave it blank or write "whatever" (or whatever).

row in which relation to column?	tilling	rotovating	green manure planting	fertilizing	bed building	set up irrigation	set up rows
weeding	after	after	after	after	after	after	after
set up rows						before
set up irrigation
bed building
fertilizing
green manure
rotovating

 

EDIT Julian Dec 3 Forget about the table, use dia and this file.

5. It would be nice to have some sample data for different yields of crops. For example, you wrote about the yield of tomatoes in your document; maybe you could put in data for another two or three crops. But this isn't really necessary at this stage, and later on we'll have easier ways of entering this data. So don't spend a lot of time on this one. But, if you could do it, it would be nice to have the absolute yield in (kilo)grams for each unit lot as defined above.

Crop	Week 1	Week 2	Week 3	Week 4	Week 5	Week 6	Week 7
Tomatoes	1kg (?)	?	?	?	?	4 kg (?)

6. Related to this, are these figures for crop yield in terms of ripeness constant over the entire planting season, or is this at least a good approximation? For example, are the above expectations for tomatoes valid independently of the week you plant them in, or are there variations? (i.e., in summer they might yield more fruit more quickly than in late autumn) OTOH, maybe it's too difficult to quantify this, and we had better just use one table as in 5. for every crop, independently of planting time. But with more precise data, you get better planning....