Artificial Intelligence
5. Optimization & Decision-making Algorithms
Lesson 2 Resource Scheduling Problem
Lesson Resource Scheduling Problem - Artificial Intelligence - ثالث ثانوي
Part 1
1. Basics of Artificial Intelligence
2. Artificial Intelligence Algorithms
3. Natural Language Processing (NPL)
Part 2
4. Image Recognition
5. Optimization & Decision-making Algorithms
Scheduling Problems
Table 5.1: Real-world applications need scheduling solutions
Single-Machine Weighted Tardiness (SMWT) Problem
Job Shop Scheduling (JSS) Problem
Using Python and Optimization to Solve the SMWT Problem
The random.randint(x,y) function is used to generate a random integer between x and y. A different
The compute_schedule_tardiness() function will be used to evaluate schedules and will serve
Itertools.Permutations() Function
Brute-Force Solver
The solver returns the best schedule, its tardiness, and the
Greedy Heuristic Solver
The following function implements an alternative heuristic that also takes into account the weights
Local Search
Local_search_solver() Function
The following function implements a random swap by simply selecting two random jobs in the given schedule that should exchange places.
The following code uses both swap strategies with the local search solver to solve the sample problem generated in the beginning of this lesson.
Comparing Solvers
Compare() Function
Describe two different strategies (swapping, inversion, shifting etc.) for the local search approach of solving the SMWT problem.
How many possible schedules (solutions) are there for an instance of the SMWT problem with 9 jobs?
You want to create a brute-force solver for the SMWT problem. Complete the following code so that the function utilizes brute force to find the optimal scheduling permutation.
You want to create a local search solver for the SMWT problem. Complete the following code so that the function utilizes local search to find the optimal scheduling permutation.
Describe how local search works.
Write your observations about the results of the the greedy solvers compared to the local search solvers, in the 30 job problem. Why do you think in the 30 job problem the brute-force solver
