Message Boards

WOLFRAM COMMUNITY

6474 Views

6 Replies

4 Total Likes

View groups...

Follow this post

Share this post:

GROUPS:

Wolfram Language Optimization

Using a list for possible values that a variable should have

Wael Al Hajailan

Posted 4 years ago

Hi, I need to assign possible values for a variable and then to chose the optimal value? Like a=[1,2,5,8] and then to find the optimal value among them according to an optimization purpose.

POSTED BY: Wael Al Hajailan

6 Replies

Sort By:

David Keith

Posted 4 years ago

There are no doubt one-line statements which can do this, but here is a hopefully understandable approach: (* objective function ) obj[x_] := (x - 5)^2 + 3 Plot[obj[x], {x, 0, 10}] ( candidate values ) xList = Range[1, 10]; ( return a trial value and its function value ) try[x_] := {x, obj[x]} ( try them all ) tryList = try /@ xList ( find the minimum ) MinimalBy[tryList, #[[2]] &] ( {{5,3}} ) ( the minimizing value is x = 5 for a min of 3 *)

There are no doubt one-line statements which can do this, but here is a hopefully understandable approach:

(* objective function *)
obj[x_] := (x - 5)^2 + 3

Plot[obj[x], {x, 0, 10}]

(* candidate values *)
xList = Range[1, 10];

(* return a trial value and its function value *)
try[x_] := {x, obj[x]}

(* try them all *)
tryList = try /@ xList

(* find the minimum *)
MinimalBy[tryList, #[[2]] &]

(* {{5,3}} *)
(* the minimizing value is x = 5 for a min of 3 *)

POSTED BY: David Keith

kritesh anand

Posted 4 years ago

Linear programming with interval coefficients is one of the tools to tackle uncertainty in mathematical programming models. This paper presents a problem solving linear programming with interval coefficients. The problem will be solved by the algorithm general method to solve linear programming with interval coefficients. Thus, the best and the worst optimum solutions can be obtained, the upper bounds and lower bounds for the optimum value of the main problems can be determined.

POSTED BY: kritesh anand

Rohit Namjoshi

Posted 4 years ago

It is not clear exactly what you are trying to do. Assuming your "optimization" computation can be expressed as a function `f` and the "optimal value" is the `Max` of that function for each of the values then e.g. a = {1, 2, 5, 8}; f[x_] := x^2; f /@ a // Max (* 64 *)