« June 2018 » S M T W T F S 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Some URLs
My Research Journal
Wednesday, 5 April 2006
Meeting with Mick
Mood:  a-ok
Now Playing: I Don't Want to Wait (Paula Cole)
Topic: Interviews
So, went and meet Mick Bromilow yesterday to discuss linear programming and software. I thought perhaps he knew about linear programming in the UK and the OU, but as it turns out he is not in the linear programming research area - so, not able to network there. But he did share some light on the M373 course since he was part of the team that developed it. According to him, M373 (which had various versions before) was developed because they considered they should have an optimization course as part of the applied mathematics degree. LP used to be taught in a level 2 course but was removed to make more room for other applied maths like mechanics.

Mick indicated that although simplex is taught there are better algorithms now available such as the interior point method which students should be taught but believed that the simplex will continue as it forms part of most linear programming curricula.

He indicated that previously in-house software was created for LP, but because the need for continuous updates a commercial software was chosen - the reason for MathCad. Also, most of the students would be familiar with MathCad since they would have used it since the entry into their OU course (not to mention OU has a licence for it) - and hence made sense to do linear programming in this way. They developed a number of worksheets templates for the students and in these the students are able to solve the problem using the matrix method rather than the tableau method for formulating the problem for solving. The course is not concerned with the students being proficient at solving the linear programming by hand, but rather being able to set up large scale problems to be solved. Hence, the MathCad just requires the student for pressing F9 to solve the problem. There is however a window in which they can see how the calculations are occurring in the background, although students are required to know the terms pivot point, pivot row they don't actually need to know these for carrying out the computer solution. However, TMA questions may require the student to examine the background calculation window to describe what might be occurring.

The templates were developed by them and the LTS team to make sure they could be readable and understandable to the student.

He also indicated that as this is a maths course they weren't too concerned about the graphics side of the LP, as that is only a 2 variable problem - but it is taught to give an understanding of how solutions are found (I'm not quite sure if this is what he actually said - but I think so ).

Posted by prejudice at 11:30 AM BST
Thursday, 16 February 2006
Making my mandatory post
Mood:  chillin'
Now Playing: Lips Like Sugar (Seal)
Topic: Interviews
so, I'm forced to make a post into this blog since, my reminder keeps popping up to say go make a post ... and I can't very well make it go away unless I do it.

Anyway, today I started my telephone interviews of the students ... well, so far I've only done one ... got one definitely booked tomorrow at 1:15 and another this afternoon after 6 ... so, that will be three definite interviews. I have a feeling that might be all that I might get - since I keep calling the other people's phone and it keeps ringing out or switched off - but I have to keep at it.

The reason I took so long to get on with these interviews was because I couldn't get a hold of the telephone recording equipment - but today when I called up Sarah it turns out there were cassette recording systems but they were trying to get me the digital systems, that is why they didn't reply my request (well at least I think so!) - but I've taken the cassette recording ones - I don't mind - I can always translate it into digital output ... if there is an audio output in the player which I could then feed into the computer - bit long - but will prefer to have it in digital format.

The one girl I've spoken to so far, says she can hardly remember what she did - but I thought she remembered pretty well. She hadn't done any of the problems etc - as I suspected. I expect people won't attempt these until their mini-project comes around which is due in mid-march. But I've been able to extol a promise from this interviewee that I can interview her again at the end of March to see how she got on with her project. Hopefully, will find out something more then. Because as of now, she has not attempted any problems, did any graphical problems or looked at Excel Solver - so couldn't get much out of it. They'll definitely have to do that for the mini-project ... so, I think I might get more rich data then.

Posted by prejudice at 5:53 PM GMT