pair & check variables

[dunyayigezerken]

Hello, I have a problem with check 2 variables. For example;

Assume that a(i) and c(i) as
a(j) c(i)
60 105
22 67
64 109
66 111
68 113

for new a(j) value the model checks that a(j)>c(i)-5 for example if new value can be 70 due 70>67-5 however if it is realized, the model use c(i) that is the result of 22. after that according to the model restrictions next a(j) can't use the result of 22. Because it was already used before, so the model can only check it with 105,109,111,113.

How can we provide this with the model? as in its algorithym, it seems that except the indices which was used before but in maths model I couldnt do ??

Thank you...

[bussieck]

I have the feeling all you describe can be modeled with binary variables. But I do not understand the problem well enough to give a concrete suggestion. For example, you describe how the new a series get developed, but you also use a and c almost as a function "the model use c(i) that is the result of 22". What happens to the c values when you select a new a. Is this all in isolation or is it a part of a bigger model. Perhaps it helps if you describe the original problem.

-Michael

[dunyayigezerken]

Hello,

Actually,I have o() values as parameter and a()=o() - e(), e() which is variable lower than 40, the new a(j) (6th a in the example) must be greater than c(i)-5 while j>i however;
every new a must use just 1 c(i). For example, if we use 67 for finding 6th a(j), 7th a(j) can not use 67 to check again. because if we use it again, it means that, for example, you have an oven and you have a product 6th a(j), 1 product to put it at second 67 however you cannot use the second 67 again for 7th a(j). But, we can use another c(i), so for 7th a(j), the model checks all c(i) which is j>i and except used c(i), in this example it is 67 for example.