Hello all,
New here, so excuse my possible dumb questions...
So I intend to try to build a timetable for our faculty- it will be a test-attempt just with a section- but after that with the entire faculty (more than 800 students)
My problem is that for a terminal year I do not know how to make the input for the clinical modules. To be more precise, we work in multiple weeks clinically orinetated modules. So, along the 14 weeks of a semester, there are 4 modules-M1,M2,M3, M4. Each of these module comprise about 15-20 hours/week but may be uneven. M1 lasts 3 weeks in a row, M2 lasts 5 weeks, M3 4 weeks and M4 2 weeks. Each M is distributed during 1 week as a 5 hours on day 1, 8 hours on day 2 and 6 hours, let's say on day 3. All of these modules are to be followed by several divisions of the year- let's say G1, G2, G3...
Who can help me?
So the first thing you must decide is:
a) do you want to enter/schedule the modules M1-M4 in a single table? if you want that, then you must enter 4*5 or 4*7 days. each activity must get a M1, M2, M3 or M4 tag; so you can place them with a preferred time constraint into the correct week.
or
b) do you want to schedule the modules M1-M4 separated in single files. So have a file only M1, one only for M2, ... . So simply do a normal week and add only activity of a single Module. So no need to care about Mx-Tags and Mx-time-constraints.
It's a bit depending on your needs. If you are well skilled in timetabling (with other tools) you can choose like you want.
I recommend to choose variant b) if you are a beginner.
(because Variant b) is easier to understand; it's easier to find bugs; it's faster to calculate; ... Disadvantages of b) are (in my opinion) minor: more files needed. So also 4 times printing/publishing. Also no constraints over the weeks.)
Thanks for the fast reply.
Most probably I made myself not sufficiently clear. M1-M4 modules have to be linked on a timeline -it may be randomly ordered, but to last for 2,4,5,2 weeks each (as explained earlier). The year that has to follow this series of modules is divided into 3 parts (G1-G3) that all have to go through these modules. And the bad thing is they all have to fit into the schedule of 1 year (in fact the last year), sharing rooms&teachers with other years....so everything has to be in a single file!
PS- I am a beginner...sorry...
Alex
Quote from: alxgudea on October 23, 2012, 06:59:16 PM
M1-M4 modules have to be linked on a timeline -it may be randomly ordered, but to last for 2,4,5,2 weeks each (as explained earlier).
No problem. Even if you must change the order later (maybe into 1 week M1 + 1 week M2 + 1 week M3 + 1 Week M4 + 1 week M1 + 3 week M2 + 4 week M3 + 1 Week M4).
That doesn't change something. Entering the data and calculating a solution is the same. So if you have a solution you can still change the order randomly like you want, because you need to tell students and teachers only the needed Mx week number for each week. So still solution a and b are correct.
Quote from: alxgudea on October 23, 2012, 06:59:16 PM
The year that has to follow this series of modules is divided into 3 parts (G1-G3) that all have to go through these modules.
No problem as long as each year has the same module at the same time. So if G1 has module M1 also G2 must have module M1. I still recommand variant b) then.
It's only difficult if the modules change randomly also in the years. So if G1 has module M1, but year G2 might have module M1, M2, M3, or M4 simultanously at the same time.
If the activities of the modules change only a bit, then you can use one of the "fortnightly" tricks: compare http://www.timetabling.de/manual/FET-manual.en.html#id_36
How did you solve that problem in the past? How did you solve that problem last year?
Quote from: alxgudea on October 23, 2012, 06:59:16 PM
....so everything has to be in a single file!
Are you sure? Several schools also have A and B weeks (so like your modules). They even change that twice a year. And all the activities of a student are linked to activities of students with previous and future years (time). But this problems can be always solved separated. Most schools just enter several files/datasets. Always separated. So they calculated an A and an B week, even they change that A and B timetabling overall 40 times in a students career.
How did you enter/solve the timetable last year?
Hi,
Again many thanks for your reply.
Imagine that all this stuff has been done manually by one man...can you imagine how nervous this man was during the very 2-3 first weeks of the semester?
I will try the solution you suggested....I will come back later for other "professional" advice.
Thanks again,
Alex
Please see a possible solution here: http://lalescu.ro/liviu/fet/forum/index.php?topic=3491.0
(a custom FET version allowing multiple-week timetabling).