Autonomous Agen Pembeli Dan Penjual Dengan Pendekatan Multi Objective Optimization Using Evolutionary Algorithm (Moea) NSGA II

Rolliawati, Dwi (2011) Autonomous Agen Pembeli Dan Penjual Dengan Pendekatan Multi Objective Optimization Using Evolutionary Algorithm (Moea) NSGA II. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 2209205208-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Game dikatakan menarik jika melibatkan permasalahan multi obyektif. Inti dari game terletak pada perilaku agennya yang dibekali kecerdasan buatan. Dalam penelitian ini, didesain perilaku agen pembeli dan penjual dengan pendekatan optimasi multi obyektif menggunakan algoritma MOEA (Multi Objective Optimization using Evolutionary Algorithm ) NSGA-II ( Non Dominat Sort Genetics Algorithms-II). Konsep multi kriteria pada agen dalam penelitian ini didasarkan teori Constraint Satisfaction Problem (CSP). State perilaku agen diatur dengan finite state machine (FSM). Agen pembeli memiliki variabel harga dan spesikasi yang berpengaruh pada faktor kepuasan agen. Agen penjual memiliki variabel harga dan laba yang berpengaruh pada fungsi join utility. Penelitian ini menghasilkan solusi optimal bagi agen yang ditampilkan bentuk grafik. Berdasarkan hasil simulasi diujikan tiga macam skenario tipe agen pembeli dan penjual yaitu agen penjual berorientasi pada laba dengan menaikkan nilai join utility (JU) sebesar dua kali lipat, agen penjual berorientasi pada pelanggan dengan menggunakan kebijakan JU normal atau menurunkan JU sebesar 0.5 dan agen pembeli yang berorientasi pada kepuasan dengan faktor kepuasan (Bij) ≥ 1 . Kestabilan hasil simulasi rata-rata dicapai setelah generasi ke-5 dengan parameter simulasi: kromosom/pop,= 200, probabilitas crossover (pc)=0.9, probabilitas mutasi (pm)=0.005, indeks distribusi crossover (ηc)=20, indeks distribusi mutasi (ηm) =20, nilai pool=pop/2 dan tour=2.
===================================================================================================================================
The game is called attractive if it is involve multi objective problems. The main of the game is based on its agent behavior which is equipped an artificial intelligent. In this research, we are designed buyer and seller agent behavior by multi objective optimization’s approach using Multi Objective Evolutionary Optimization (MOEA) NSGA-II. The concept of multi criteria for agent in this research is based on Constraint Satisfaction Problem’s theory. The agent behavior’s state is based on finite state machine (FSM). The buyer agent has price and specification variable which is reacted in satisfaction factor of agent. The seller agent has price and profit variable which is took effect in join utility of agent. The research is produce optimal solution for agent which is performed by graphically. Based on simulation result, we are generated 3 kinds of scenarios of buyer and seller agent typist, such as seller agent is profit oriented with put up the value of join utility (JU) amount of twice from the buyer’s function, the seller agent is customer oriented with normally JU or put down the join utility (JU) =0.5 from the buyer function and the last is the buyer agent that satisfaction oriented with satisfaction factor (Bij) ≥ 1. Stability results of simulation is evenly attained after the fifth generation with simulation parameter’s: chromosome/pop=200, crossover probability (pc)=0.9, mutation probability (pm)=0.005, index of distribution crossover (ηc)=20, index of distribution mutation (ηm) =20, value of pool=pop/2 and number of tour=2.

Item Type:	Thesis (Masters)
Additional Information:	RTE 006.3 Rol a 2011 3100012045923
Uncontrolled Keywords:	game, multi obyektif, agen, pembeli, penjual, CSP, FSM, simulasi, multi objective, agent, buyer, seller, MOEA NSGA-II, simulation, join utility
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Divisions:	Faculty of Industrial Technology > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Anis Wulandari
Date Deposited:	07 Jan 2026 04:51
Last Modified:	07 Jan 2026 04:51
URI:	http://repository.its.ac.id/id/eprint/129329

Actions (login required)

View Item