ALGORITHM DEVELOPMENT OF NEURAL NETWORK ALGORITHM (NNA) TO SOLVE REAL OPTIMIZATION PROBLEMS

• Main Author: Sipayung, Michael
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/70354
• Institution: Institut Teknologi Bandung

Inspired by Artificial Neural Network (ANN), a metaheuristics optimization algorithm which then so-called Neural Network Algorithm (NNA) is utilized to solve optimization problems. In addition to unconstrained problem, NNA is also can be modified to find the constrained optimization problem although only the continuous variable can be obtained in one program execution. In this thesis, to find possible solution of optimization problem which is involve the integer , the original NNA can be modified using rounding techniques. By using this technique, NNA can obtain the optimization problem which is involve the integer. Furthermore, the modification of NNA algorithm is also can be utilized to solve real optimization problems. Finally, after doing test to all the benchmark functions, NNA with it’s modification has been tested successfully such us constrained optimization problems, finding the solution of real optimization problems.