AbSTRAkPenelitian ini mengembangkan algoritma Differential Evolution (DE) untuk menyelesaikan permasalahan penjadwalan flow shop m-mesin dengan mempertimbangkan dua obyektif yaitu makespan dan total flow time. Pengembangan algoritma DE dilakukan dengan menambahkan adaptive parameters pada tiap generasi, menggunakan strategi local search untuk meningkatkan kualitas solusi yang dihasilkan dan memodifikasi crossover untuk mengurangi waktu komputasi. Hasil penelitian ini menunjukkan bahwa algoritma DE yang diusulkan memiliki kinerja yang lebih baik dibandingkan dengan algoritma DE murni, algoritma Genetika (GA), dan pada kasus tertentu juga memiliki kinerja yang lebih baik dibandingkan algoritma Multi-Objective Ant Colony System (MOCSA). kata kunci: flow shop scheduling, multi-objective, makespan, total flow time, differential evolution AbSTRACT This research focuses on the development of Differential Evolution(DE) algorithmto solve m-machine flow shop scheduling problems with respect to both makespan and total flow time. Development of DE algorithm is done by modifyingthe adaptive parameter determination procedure in order to change the value of adaptive parameters in each generation, adding local search strategy to the algorithm in order to improve the quality of the resulting solutions, as ewell as modifying the crossover in order to reduce computation time. The result indicates that the proposed DE algorithm has proven to be better than the original DE algorithm, Genetic Algorithm (GA), and for certain cases it also out performs Multi-Objective Ant Colony System Algorithm (MOCSA).

No abstract

This study presenting the result of forecasting sales of Hex Nuts between the Trend Linear Line (TLL) method and Monte Carlo Simulation. To determine the appropriate method, the Mean Average Percentage Error (MAPE) is used to evaluate theerror rate. We find that the Monte Carlo simulation outperforms the TTL method, where the MAPE value of the Monte Carlo simulation is 7,61%. Based on the result, the Monte Carlo simulation is the appropriate method to forecast the sales rate of Hex Nuts in the PT. KMS.  

Abstract-