Articles
Penerapan Algoritma Genetika untuk Optimasi Vehicle Routing Problem with Time Window (VRPTW) Studi Kasus Air Minum Kemasan
Dita Sundarningsih;
Wayan Firdaus Mahmudy;
Sutrisno Sutrisno
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 1 No 2 (2017): Februari 2017
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (655.267 KB)
Salah satu permasalahannya dalam bidang optimasi yaitu penentuan rute distribusi air minum kemasan. Air merupakan salah satu kebutuhan pokok bagi manuasia, sehingga banyak sekali permintaan untuk pemasokan air minum kemasan . Penentuan rute terpendek sangat penting karena pengiriman barang harus dilakukan dengan singkat dan tepat waktu dengan memaksimalkan penggunaan alat transportasi untuk mengurangi biaya transportasi. Vehicle Routing Problem (VRP) cenderung menyelsaikan permasalahan dengan meminimalkan biaya yang direpresentasikan oleh total jarak tempuh dan jumlah kendaraan yang digunakan. Oleh karena itu untuk menyelsaikan masalah lebih tepat menggunkan (Vehicle Routing Problem With Time Window) VRPTW, dengan tujuan menentukan optimasi rute yang dipengaruhi dengan Time window. Time window yang merupakan waktu pelayanan khusus yang disediakan oleh pelanggan. Algoritma Genetika merupakan salah satu algoritma yang dapat diterapkan untuk menyelesaikan Optimasi Distribusi Air Minum Kemasan dengan mendapatkan rute terbaik. Pencarian Solusi dilakukan dengan mengkombinasikan kromosom kemudian diproses dengan operator genetika (crossover, mutasi dan seleksi) dengan menginisialisasi parameter genetika (ukuran Populasi, probabilitas crossover, probabilitas mutasi dan jumlah generasi). Dari hasil pengujian diperoleh hasil terbaik dengan nilai fitness tertinggi pada ukuran populasi 100, jumlah generasi 2500 nilai probabilitasi crossover 0,3 dan probabilitas mutasi 0,7.
Optimasi Keanggotaan Fuzzy Tsukamoto Menggunakan Algoritma Genetika pada Penentuan Prioritas Penerima Zakat
Alfiani Fitri;
Wayan Firdaus Mahmudy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 1 No 2 (2017): Februari 2017
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1072.103 KB)
Before distributing zakat, required the selection data of recepients by considering some of its criteria because the scholarship must be received by people who are truly in need. The amount of data along with the various criteria that lead to the selection time to produce a long lasting decision. The system of priority SOS scholarship recipients are expected to produce a recommendation candidate who is suitable to receive the zakat with a relatively short time. In some studies mention that the use of fuzzy Tsukamoto methods sometimes does not necessarily produce the optimal final result. In this study, the accuracy of the test results by using fuzzy Tsukamoto alone is 0.725. One of the technique that can be used to provide a more optimal final result is the optimization of limited membership functions using genetic algorithms. The chromosome representation which is used is real-coded, crossover technique using a one-cut-point crossover, mutation techniques using random mutation and chromosomal repair using a hill climbing algorithm. The calculation result of the accuracy system by using the Spearman correlation in the optimization of Tsukamoto fuzzy membership using genetic algorithms with chromosomal repair mechanism is 0.986. While the accuracy test results of the optimization system of Tsukamoto fuzzy membership by using a genetic algorithm without chromosomal repair mechanism is 0.845. The accuracy results shows the increase of the accuracy value with the function optimization of Tsukamoto fuzzy membership by using the genetic algorithms and chromosome repairs with hill climbing.
Pemanfaatan Algoritma Genetika Untuk Optimasi 0/1 Multi-Dimensional Knapsack Problem Dalam Pendistribusian Produk (Studi Kasus UD.TOSA)
Ryan Iriany;
Agus Wahyu Widodo;
Wayan Firdaus Mahmudy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 1 No 4 (2017): April 2017
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1062.046 KB)
As a distributor company, the cost of distribution is very influential on the benefits to be obtained UD.TOSA. The affect of distribution cost is the distance distribution. Besides affected by distance, cost is also influenced by the frequency of maintenance vehicles used in the distribution process. The more frequent occurrence of damage to the vehicle, it will increase the frequency of maintenance so that adds to the cost of distribution. One cause damage to the vehicle is because vehicles are often excess payload (overtonase). Excessive loads can also increase the potential for accidents that could result in damage to the product as well as the vehicle itself. This will result in reduced profits obtained. Products are distributed and used vehicles have their respective characteristics. Each vehicle has a limited capacity, so not all products can be loaded, the distributor can perform any combination of products that should be loaded in order to maximize cargo volume without exceeding the capacity of the vehicle. The combination of products in the distribution process is a complex combinatorial problems, problems of this combination into the multi-dimensional knapsack problem (MKDP). Utilization of genetic algorithms in multi-dimensional knapsack problem is to perform such optimization of capacity in the distribution process.The algorithm parameters used in this study is a population of 200, the generation of 100, cr by 0.9 and 0.1 mr. Excess load on the solutions produced by the system is equal to 0% of the maximum load capacity of the vehicle. Solutions generated by the system can be ensured not exceed the capacity, both of maximum space vehicles as well as the maximum load of the vehicle. It can reduce the risk of damage to the vehicle so that the frequency of maintenance is not too often.
Optimasi Komposisi Pakan Untuk Penggemukan Sapi Potong Menggunakan Algoritma Genetika
Muhammad Noor Taufiq;
Candra Dewi;
Wayan Firdaus Mahmudy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 1 No 7 (2017): Juli 2017
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1160.171 KB)
One of the problems that exist in Indonesia is not share of the number between the demand for beef cattle with the number of local beef cattle production. This caused by the increase in the number of people in Indonesia. It makes Indonesia have large enough dependencies to import cows from abroad to fulfill the need of Indonesian people. This study tries to implement the genetic algorithm to creating a qualified mixed ration at the reasonable cost. This study is expected to be able to increase the number of local beef cattle production to fulfill the need of Indonesian people. The representation used in this study is real code in which each chromosome initialize feed materials which used. The mutation method is the random mutation, and the selection method is elitism. The result of this study found the optimal parameter at 900 population, 800 generation and the combination of cr and mr as many as 0.9 and 0 with the highest fitness is 0,6266. The result obtained in the form of ration composition recommendation at minimal cost as daily based of the nutritional need of beef cattle
Optimasi Model Himpunan Keanggotaan Fuzzy Menggunakan Algoritme Evolution Strategies (Pada Data Diagnosis Penyakit Sapi Potong)
Dyan Putri Mahardika;
Wayan Firdaus Mahmudy;
Agus Wahyu Widodo
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 1 No 8 (2017): Agustus 2017
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1173.825 KB)
The increasing number of daily consumption of beef, it is need to monitoring the availability of good-quality beef, which is free from germs. In this research will be created a system that can be solve the subjective, unclear and vague problem, that is cattle disease diagnose, the existence of subjective information about the symptoms of the disease, then the diagnosis can be done using a fuzzy inference system. To obtaining the high accuracy, the set of fuzzy membership needs to be determined precisely. To get the best solution or in this case is the set of fuzzy membership can be done using Evolution Strategies method. The accuracy of the system is obtained by matching the results of the diagnosis and the diagnosis expert system using 51 generates test data accuracy rate of 95.49% using fuzzy. After optimization of the fuzzy set membership using evolution strategies (ES) produces better accuracy rate is 97.45% with the best parameters obtained ES is the population size of 80, offspring size of 10μ, and the amount of generation by 50.
Optimasi Komposisi Makanan untuk Penderita Hipertensi Menggunakan Variable Neighborhood Search
Aprilia Nur Fauziyah;
Wayan Firdaus Mahmudy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 1 No 9 (2017): September 2017
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1092.854 KB)
Hypertension is a major symptom that cause other diseases appear such as noncommunicable diseases, cancer, and diabetes. One of healthy life effort in the patients with hypertension is a healthy food that attentions level of salt in foods. The food composition with attentions level of salt and cost of the food can use Variable Neighborhood Search (VNS) algorithm. The data which are used for this study are 103 data food that consists of staple foods, vegetables, vegetable sources, animal sources and fruits. VNS have 3 stages after generate initial solution, there are shaking, local search and move or not. The results of Kmax and local search's testing show that the average of highest fitness is 0.516848 with Kmax = 13 and the highest average fitness is 0.524301 with 2000 iterations. Kmax and local search iterations that too much not guarantee a solution with great fitness is obtained but many Kmax numbers or local search iterations can give wider search opportunities. The result of this study contains composition of foods with nutrients which are close to the needs of hypertension patients with attentions the natrium and minimal cost within a day.
Hibridisasi Algoritme Genetika dan Simulated Annealing untuk Optimasi Multi-Trip Vehicle Routing Problem with Time Windows (Studi Kasus: Pariwisata Kabupaten Banyuwangi)
Amalia Kartika Ariyani;
Wayan Firdaus Mahmudy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 1 No 12 (2017): Desember 2017
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (663.376 KB)
Along with the rapidly excalation in the number of tourists as well as tourism which are located in Banyuwangi, problems arise from the tourists. With limited time, they would like to visit as many destinations without wasting time in traveling. However, another problem happened when the visitor and each visited destination has time windows or visiting time which has been established. From this issue, it needs a solution to solve the problems regarding with the tourism scheduling with the optimal time based on time windows from the tourists and the destinations. Generally, these problems called Vehicle Routing Problem with Time Windows (VRPTW). VRPTW problems can be solved with several optimization methods, including using hybridizing genetic algorithm and simulated annealing. Genetic algorithms (GA) will be used to explore the global optimum problem, while Simulated annealing (SA) will be used to exploit a local optimum problem. The fitness value approach optimal i.e. of 1.0616 obtained by using parameters of GA-SA on the size of population of 400, amount of generation by 800, combination of cr and the mr of 0.3 and 0.1, initial temperature of 0.9, cooling factor of 0.9, and the coefficient of acceptance of a new solution amounting to 200. Using these parameter values hybrid GA and SA outperforms separated GA and SA.
Optimasi Kebutuhan Gizi untuk Balita Menggunakan Hybrid Algoritma Genetika dan Simulated Annealing
Fitri Anggarsari;
Wayan Firdaus Mahmudy;
Candra Dewi
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 1 No 12 (2017): Desember 2017
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (768.359 KB)
The nutritional state of a person is basically influenced by dietary behavior so that the quantity and quality of the food and beverages consumed has an impact on a person health. Balanced nutrition plays an important role in the growth, physical development, and intelligence of all people, including toddlers, children, and adults. Nutrition in toddlers should be considered because at that time they growth and develope so rapid and prone to occur bad things such as infections that can cause chronic illness, obesity and even death. In this research, we implement hybrid genetic algorithm and simulated annealing to know optimize nutrition requirement on food composition for toddlers. There are two segments of the chromosomal representation that is used in this research, the first segment uses the integer number and the second segment uses the real code number. We use extended intermediate crossover method and random mutation method for the reproduction process. The test resulted in the highest average fitness value of 0.10106 with the best parameters are population = 100, generations = 50, combination between Cr and Mr = 0.8 and 0.3, value of alpha = 0.8, value of T0 = 2 and value of Tn = 0.2. The results of this study is recommendations of foodstuffs according to the nutritional needs that approached the actual needs of the toddlers by considering the weight of food and the minimum price in one day.
Optimasi Menu Makanan Bagi Pasien Gagal Ginjal Menggunakan Algoritme Lebah
Yeni Herawati;
Wayan Firdaus Mahmudy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 4 (2018): April 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (753.615 KB)
Kidney failure is a chronic metabolic disease that becomes a health problem in the world. Patients with kidney failure should have a special diet to control their diet and keep the kidney damage from getting worse. Diet to help patients improve eating and sports habits to gain better metabolic control. Patients should limit what foods can be consumed. This causes the patient to really pay attention to the food menu. To solve the case requires a bee algorithm. This study used 132 lists of foods with nutritional content (protein, carbohydrates and fats). The algorithm process uses an integer representation of length 15 which the resulting number represents the food index. From the test, we get the optimal parameter of 100 individuals with the average fitness value of 0.107823, 100 iterations with the average fitness value of 0.103334. The results of this study are the composition of the food menu in a day (breakfast, lunch and dinner) at a minimal cost.
Optimasi Penataan Barang pada Proses Distribusi Menggunakan Algoritme Evolution Strategies
Rani Kurnia;
Wayan Firdaus Mahmudy
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 5 (2018): Mei 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1253.491 KB)
Distribution is the one of important processes in industrial activities such as goods's distribution. But unfortunately, this process is one of processes that spend much expense. Then it be required a great management to minimize costs for that's process. One of them is to optimization the goods organize management. Optimization problem of goods organize management can be solved using evolution strategies algorithm that known as a simple algorithm that can be solved a complex problems. In this case, evolution strategies algorithm start with raising the random individu, then mutation process, then do the goods organize management process to get the fitness result and close with selection process to get the great individu. Based on the test of goods organize management that have been process with 50 goods, it get the optimal parameter consist of population 60, lambda 18, offspring 18μ, and generation 15 with the fitness's average is 2,404. The final result of this experiment is a scheme of the optimal solution for goods organize management in vehicle.