Computer Science

– Assessment or evaluation is an element of the student and learning. The assessment gives control of value at each period of the instructing learning process. Javanese script writing is a part of Javanese Studies, which taught in elementary to junior high school. Mostly, the teacher has no standardization to scoring how well the Javanese Character handwriting. Optical character recognition (OCR) is a character automatic computer recognition process in optically scanned and digitized manuscript. In this research, we will implement character recognition approach for electronic document as a tool for evaluation measurement on Javanese character writing learning. OCR system would recognized and interpreted Javanese script handwriting to be suitable meaning and matched it with the teacher's answer key. In a nutshell, if OCR system could not or incorrectly to recognized the specific character handwriting which expected by the teacher, the handwriting can be concluded as incorrect or poor handwriting. The experiment result shows that our proposed method can generate 85% accuration rate. Accuration rate based on how much the system able to recognize which handwriting that classified as correct and proper handwriting, and poor or incorrect handwriting which deliberately inputted. By applying our method, the teacher can measure student Javanese script handwriting of with 85% confidence level of the score.

Abstrak Pulau Bali merupakan tempat impian bagi wisatawan dimana terdapat ratusan bahkan ribuan destinasi wisata yang ditawarkan[1]. Waktu wisatawan yang singkat dan banyaknya destinasi wisata yang ingin dikunjungi, membuat wisatawan harus menjadwalkan perjalanan wisatanya seefektif mungkin. Pada kenyataannya, setiap destinasi wisata mempunyai waktu buka-tutup atau waktu terbaik untuk dikunjungi. Pemakaian konsep TSP konvensional akan menjadi kurang tepat jika di implementasikan pada kondisi demikian. Salah satu bentuk pengembangan TSP yang lebih rumit dengan melibatkan dua variabel adalah TSP-TW yaitu pencarian rute optimal dengan menambahkan variabel waktu yang harus diperhatikan[3]. Pada penelitian ini, akan digunakan algoritma genetika pada kasus TSP-TW untuk menghasilkan jadwal perjalanan paling optimal yaitu dengan rute terpendek dan perjalanan tepat waktu pada wisata pada Pulau Bali. Berdasar hasil pengujian, parameter yang optimal untuk optimasi TSP-TW menggunakan Algoritma genetika pada kasus penjadwalan perjalanan wisata pada Pulau Bali yaitu metode seleksi yang di pakai adalah Elitis, nilai Crossover Rate (Cr) adalah 0,05, nilai Mutation Rate (Mr) adalah 0,35, jumlah generasi adalah 1750 generasi dan jumlah populasi adalah 100 populasi. Abstract The Bali Island is a dream place for tourists where there are hundreds or even thousands of tourist destinations that offered[1]. In the reality, the tourists don't have much time enough to visit the abundant tourism destination so that the travelers should make tourism travel schedule as effective as possible. In fact, each destination has open-close time or the best time to visit. The use of the conventional TSP concept would be unreliable if implemented in such conditions. One variant of TSP concept which more complicated by involving two variables is TSP-TW where this concept will search the optimal route by adding a time variable that must be considered [3]. In this research, we will use a genetic algorithm on TSP-TW case for optimizing the travel scheduling on the Bali Island. Based on the test results, the optimal parameters for the TSP-TW optimization using genetic algorithms on travel scheduling to the Bali Island which Elitist as the selection method , crossover rate value is 0.05, crossover rate value is 0.35, the generation size is 1750, and the population size is 100. 1. PENDAHULUAN Pulau Bali merupakan tempat impian bagi wisatawan lokal maupun mancanegara. Pesona objek wisata baik berupa keindahan alam, budaya dan adat istiadat yang kental membuat Bali menjadi destinasi wisata unggulan di Indonesia. Terdapat lebih dari ratusan bahkan ribuan destinasi wisata yang ditawarkan oleh Pulau Bali. Banyak paket pariwisata yang ditawarkan oleh hotel tempat menginap atau agen perjalan [1]. Setiap paket perjalanan telah mempunyai jadwal dan destinasi yang telah ditentukan sebelumnya. Tidak sedikit wisatawan yang kurang menyukai paket yang disediakan oleh hotel atau agen. Hal ini terjadi karena ketidaksesuaian jadwal yang telah ditentukan dengan keinginan para wisatawan. Banyak wisatawan memilih untuk berwisata " backpacker " dengan menentukan destinasi-destinasi wisata sesuai keinginan mereka sendiri. Pada umumnya, semua wisatawan tidak mengetahui jarak antar destinasi wisata dan wisatawan hanya mempunyai waktu yang singkat untuk menikmati liburannya namun juga ingin mengunjungi banyak destinasi wisata. Waktu yang singkat dan banyaknya destinasi wisata yang ingin dikunjungi, membuat wisatawan harus menjadwalkan perjalanan wisatanya seefektif mungkin.

Poverty level reduction is the most important part in the development policy in Indonesia. Various types of government policies have been conducted as an effort to reduce poverty, one of which is Bantuan Langsung Tunai (BLT) program. However, the distribution of the BLT program frequently not right on target. The process of determining priority of poor households is quite complicated because it involves many parameters that must be taken into account. This study uses Tsukamoto Fuzzy to assist in determining the priority of poor households. The goal is to form a ranking so the sequence of households who eligible to get assistance can be known. Criteria used in the calculation were based on BPS provisions concerning the poor household criteria. The test results using Spearman Correlation obtained correlation value is 0.7776, which indicates that there is a high association between the ranking, which generated by the system and the ranking results from BPS calculation. The final results obtained is the ranking of poor households of 30 data were used. ABSTRAK Pengentasan kemiskinan merupakan bagian paling penting dalam kebijakan pembangunan di Indonesia. Berbagai jenis kebijakan pemerintah telah dilakukan sebagai upaya untuk menurunkan angka kemiskinan, salah satunya adalah dengan adanya program Bantuan Langsung Tunai (BLT). Namun pemberian program BLT tersebut sering kali tidak tepat sasaran. Proses penentuan prioritas rumah tangga miskin ini memang cukup rumit karena melibatkan banyak parameter yang harus diperhitungkan. Penelitian ini menggunakan Fuzzy Tsukamoto untuk membantu dalam menentukan prioritas rumah tangga miskin. Tujuannya adalah membentuk sebuah peringkat rumah tangga miskin sehingga dapat diketahui urutan rumah tangga yang berhak untuk mendapatkan bantuan. Pemilihan kriteria yang digunakan dalam perhitungan didasarkan pada ketentuan BPS mengenai kriteria rumah tangga miskin. Dari hasil pengujian menggunakan Spearman Correlation, didapatkan nilai korelasi sebesar 0.7776, yang menunjukkan bahwa ada keterkaitan yang tinggi antara peringkat yang dihasilkan oleh sistem dengan peringkat hasil perhitungan BPS. Hasil akhir yang didapatkan berupa peringkat rumah tangga miskin dari 30 data rumah tangga yang digunakan.

Route scheduling is a quite complicated process because it involves some determinant factors. Several methods have been used to help resolve the NP-hard problems. This research uses genetic algorithm to assist in optimizing ship scheduling, that where there are several ports to be visited by some ships. The goal is to divide the ship to go to a specific port so that each port is only visited by one ship to minimize the total distance of all ships. The computational experiment produces optimal parameters such as the number of popsize is 30, the number of generations is 100, crossover rate value is 0.3 and mutation rate value is 0.7. The final result is an optimal ship route by minimizing the distance of each ship. 1. Introduction In the area of communication and transportation, the problems concerning the route and scheduling is an area of extensive research. The stable economic growth in trade resulting in an increased need for transport of goods. There are four types of modes of transport are often used for delivery of goods, namely trucks, trains, planes, and ships. Indonesia with more than 18,000 islands, is the largest archipelago in the world [1]. With so many islands in Indonesia, the manufacturers often choose to use the ship in order to send goods production between islands.Most ships are used to deliver the goods assign ships to the weekly frequency for long-distance shipping services. As a result, the goods to be sent must be stored in warehouses or in port beforehand, causing increased costs for storage. To reduce these costs, plan of ship service of goods shippers is expected to be optimized in terms of schedules and allocation of ships on some routes [2]. Fuel cost has the highest calculation proportion among the various operational costs of sea shipping. High fuel consumption gives adverse impact on the environment because it produces CO 2 and NO x emissions that cause air pollution and climate alteration [3] [4].Therefore, ship's route scheduling is important to be taken to minimize the travel distance so that the fuel consumption can be as little as possible. In the previous research, genetic algorithm is used to determine the allocation of berths for container ships in the harbor. The purpose of this allocation planning is to minimize the amount of time service for each ship [5]. A similar study conducted to optimize the scheduling of the train route using genetic algorithms with artificial neural networks with the aim of minimizing the delay time [6]. Each gene on chromosome describes a train that is assigned to a route which is represented by the value of its genes [7].Genetic algorithm is also used to optimize route of stacker in automatic warehouse because stacker can not increase its speed so it takes much time in the process of transportation goods [8]. In this study, genetic algorithm is used to optimize the ship's routescheduling because genetic algorithm have a powerto produce good solutionsfor somereal complex problems [9] [10]. This paper presents a model to solve the problem of scheduling a port visit to minimize the distance so that the fuel consumption is reduced. The proposed solution is trying to save fuel consumption of each ship along the way to visit all the ports.

The highest component in the production cost of the poultry industry is feed cost. The formation of an efficient feed composition is needed because of the increasing price of feed ingredients. Several types of software have been developed to help determine the feed composition, but the price of commercial feed formulation software is quite expensive for most organizations. Hybrid adaptive genetic algorithm and Simulated Annealing were used to calculate poultry feed formulations. This algorithm used a change mechanism of the control parameter in genetic algorithm adaptively to get better results. Simulated Annealing was applied to avoid a local optimum solution produced by the genetic algorithm. The results showed that hybrid adaptive genetic algorithm and Simulated Annealing is better than the classical genetic algorithm.

Determination of the fodder composition is a difficult process because it should simultaneously consider several constraints, such as minimizing the total cost of feed ingredients and maximizing the nutrient needs required by livestock. This study uses a modified genetic algorithm to solve the problem in order to obtain better results. The modification is done by applying numerical methods in generating an initial population of the genetic algorithm. Testing results show that the optimal parameters that can be used to produce the optimal solution are as follows: population size (popsize) is 300, generation number is 400, crossover rate (cr) value is 0.2, and mutation rate (mr) value is 0.6. The modified genetic algorithm provides an average fitness value of 0.142357, while the classical genetic algorithm provides an average fitness value of 0.094354. With additional computational time equal to 110 ms, the use of modified genetic algorithm offered has proven to provide a better result, with a higher fitness value compared with classical genetic algorithm.

Restoran merupakan salah satu jenis usaha yang mementingkan kualitas pelayanannya. Namun dalam kegiatan operasionalnya, masalah keterlambatan pelayanan sering muncul karena kurangnya koordinasi yang baik antar bagian. Salah satu faktornya adalah jarak antara dapur dan ruang makan. Sistem pengelolaan restoran terintegrasi dapat dibangun untuk mengurangi masalah tersebut. Namun pada penggunaannya, sistem harus memiliki kualitas yang baik agar dapat mendukung jalannya kegiatan operasional restoran. Sebab kualitas yang buruk pada sistem dapat berpotensi mengakibatkan error sehingga pihak yang menggunakannya dapat mengalami kerugian finansial. Penelitian ini membahas tentang pembangunan sistem pengelolaan restoran terintegrasi yang memenuhi standar karakteristik utama model kualitas ISO 9126. ISO 9126 memiliki enam karakteristik utama yaitu functionality, reliability, usability, efficiency, maintainability, dan portability. Functionality diuji dengan menggunakan pengujian validasi terhadap seluruh kebutuhan fungsional, reliability diuji dengan menggunakan tool WAPT (Web Application Performance Tools) yang fokus pada ketersediaan server dengan sessions, pages, dan hits sebagai parameternya, usability diuji dengan menggunakan model kuesioner SUS (System Usability Scale) yang disebar kepada lima belas responden, efficiency juga diuji dengan menggunakan tool WAPT namun fokus pada waktu respon server, maintainability diuji dengan menggunakan persamaan MI (Maintainability Index), dan portability diuji dengan menggunakan tool BrowseEmAll yang fokus pada kemampuan sistem untuk dapat dioperasikan di web browser yang berbeda. Berdasarkan hasil pengujian, sistem telah memenuhi standar karakteristik utama ISO 9126 yaitu dengan functionality sebesar 1, reliability sebesar 100%, usability sebesar 68,5, efficiency sebesar 0,78 detik, maintainability sebesar 66,62597, dan hasil portability menunjukkan tidak ditemukannya error maupun warning di tujuh jenis desktop browser dan lima jenis mobile browser. Selain itu, hasil pengujian integrasi menunjukkan bahwa seluruh proses yang melibatkan interaksi antar kelas telah terintegrasi dengan baik.