Hasil Pencarian  ::  Simpan CSV :: Kembali

"Besarnya kebutuhan akan kemasan kayu dalam kegiatan perdagangan menuntut produsen kemasan kayu untuk mengembangkan sistem penjadwalan produksi yang optimal agar dapat memenuhi permintaan pasar. Penelitian ini membahas optimasi penjadwalan produksi kemasan kayu, khususnya pallet, dengan system job shop melalui penerapan algoritma Differential Evolution. Prinsip algoritma DE sesuai dengan analogi evolusi biologi yang terdiri dari proses inisialisasi populasi, proses mutasi, proses pindah silang, dan proses seleksi. Tujuan dari penelitian ini adalah untuk meminimumkan total biaya (dalam satutan waktu) yang timbul sebagai akibat dari keterlambatan dalam proses produksi kemasan kayu, khususnya pallet yang dipergunakan untuk mengemas barang-barang ekspor. Penelitian dilakukan melalui studi kasus dengan mengamati proses produksi pallet pada suatu produsen pallet. Hasil dari penelitian ini adalah diperolehnya usulan penjadwalan produksi dengan penurunan total biaya keterlambatan sebesar 25,22%, (dari 23590 menit menjadi 17640 menit) serta penurunan pada kriteria lainnya seperti jumlah pesanan yang terlambat, total waktu keterlambatan, dan waktu penyelesaian. Simulasi penambahan kapasitas menunjukkan bahwa penambahan jumlah mesin pada stasiun serut dan potong dapat menghilangkan keterlambatan dan meningkatkan output hingga 95%.

---

The ever increasing need of wooden packaging , whether for international as well as inter-islands transportation of traded goods, has required the producers of wooden packaging to develop an optimized production scheduling system to fulfill the market demand. This research studies the optimization of the job shop production scheduling system of wooden packaging, particularly pallet, through the application of the Differential Evolution (DE) algorithm. The principles of the DE algorithm is in line with the biological evolution analogy which consists of the initialization of population, mutation, crossover, and selection processes. The DE algorithm has some excellent features namely the concept is simple, easy to apply, quick in producing solutions, and robust. The objektif of this research is to minimize the total cost (in time unit) resulting from the delay in the production process of wooden packaging, particularly pallet.

This research is a case-study which is carried-out by observing pallet production process at a wooden packaging manufacturer. The company produces wood packaging, mainly pallet, which are used for the packaging of exported goods. This research leads to a recommendation on the wooden packaging production schedule with more efficient cost. The total cost resulting from the delay in the production is reduced by 25,22% (from 23590 minutes to 17640 minutes). There are also reductions in the total of tardy job, total of tardiness, and makespan. A simulation of work station capacity upgrade also shows that the addition of the number of machineries in planing and sawing station will eliminate delay and increase output up to 95%."

"Penggunaan dispatching rule dalam menentukan urutan pengerjaan produk dalam flexible manufacturing system dapat memberikan solusi yang kurang baik apabila dispatching rule yang digunakan tidak sesuai. Pada tesis ini akan diperlihatkan perbandingan penggunaan beberapa dispatching rule baku dalam menentukan jadwal produksi, seperti First Come First Served (FCFS), Shortest Processing Time (SPT), Longest Processing Time (LPT), dan Earliest Due Date (EDD) serta sebuah usulan yang mempertimbangkan waktu terbuang untuk menunggu urutan pengerjaan produk pada mesin yang sama dalam waktu yang bersamaan dengan menggunakan alat bantu simulasi model. Pada tesis ini juga akan diciptakan model FMS dalam perangkat lunak, untuk kemudian disimulasikan menurut dispatching rules masing-masing, sehingga waktu penyelesaian masing-masing produk dan tardiness yang tercatat dalam hasil simulasi dapat dianalisa dan kemudian dibandingkan agar dapat dipilih untuk mendapatkan dispatching rules yang sesuai dan dapat memberikan kinerja yang paling baik. Perencanaan dispatching rule yang sesuai diharapkan dapat mempersingkat waktu penyelesaian tanpa melanggar tenggat waktu. Hasil yang didapatkan dalam penelitian ini menunjukkan bahwa dispatching rules yang sesuai merupakan kombinasi dispatching rule EDD dengan tetap mempertimbangkan waktu waktu terbuang untuk menunggu urutan pengerjaan produk pada mesin yang sama dalam waktu yang bersamaan melalui tenggat waktu minimum yang dimiliki masing-masing kombinasi urutan pengerjaan produk.

---

Dispatching rules, which commonly used in determining job execution sequences, should be well planned in order to avoid bad solution caused by the usage of improper dispatching rules. In this thesis, the comparison of several dispatching rules is made to establish a compact production schedule on an FMS using a model simulation, such as First Come First Served (FCFS), Shortest Processing Time (SPT), Longest Processing Time (LPT), Earliest Due Date (EDD) and a proposal to consider the wasted time caused in waiting execution on a same machine at a same time.

In this thesis, a model of FMS environment will be built on a software, and sequences which were produced on each dispatching rules, will be simulated based on the model created, for later the recorded value of completion time and tardiness of each product could be analyzed and compared further to gain a proper dispatching rule in giving the best performance. The proper dispatching rules planning will improve the completion time performance without violating the due dates.

The result from the case on this thesis will show that the proper dispatching rule is a combination between EDD and the proposal of considering the wasted waiting time with the concern of minimum due dates attribute owned by each product execution."

"This book presents models and algorithms for complex scheduling problems. Besides resource-constrained project scheduling problems with applications also job-shop problems with flexible machines, transportation or limited buffers are discussed. Discrete optimization methods like linear and integer programming, constraint propagation techniques, shortest path and network flow algorithms, branch-and-bound methods, local search and genetic algorithms, and dynamic programming are presented. They are used in exact or heuristic procedures to solve theintroduced complex scheduling problems. Furthermore, methods for calculating lower bounds are described. Most algorithms are formulated in detail and illustrated with examples."

"The financial results of any manufacturing company can be dramatically impacted by the repetitive decisions required to control a complex production network be it a network of machines in a factory; a network of factories in a company; or a network of companies in a supply chain. Decision policies for production networks presents recent convergent research on developing policies for operating production networks including details of practical control and decision techniques which can be applied to improve the effectiveness and economic efficiency of production networks worldwide. ;"

"The topic of this book is known as dynamic scheduling, and is used to refer to three dimensions of project management and scheduling : the construction of a baseline schedule and the analysis of a project schedule’s risk as preparation of the project control phase during project progress. This dynamic scheduling point of view implicitly assumes that the usability of a project’s baseline schedule is rather limited and only acts as a point of reference in the project life cycle. Consequently, a project schedule should especially be considered as nothing more than a predictive model that can be used for resource efficiency calculations, time and cost risk analyses, project tracking and performance measurement, and so on."

"Masalah penjadwalan batching machines adalah masalah kombinatorial untuk menyusun produk-produk sedemikian sehingga lama pengolahannya menjadi minimum. Produk-produk memiliki ukuran dan lama proses produksi yang berbeda-beda yang akan dikelompokkan dalam batches. Jumlah ukuran produk dalam suatu batch tidak boleh melebihi kapasitas mesin. Lama proses batch adalah lama proses produk terpanjang dalam batch tersebut. Mesin yang sedang berkerja tidak diperkenankan diganggu hingga semua produk didalamnya selesai diproses.Masalah ini akan diselesaikan dengan algoritma improved ant colony optimization menggunakan metropolis criterion untuk menghindari solusi konvergen yang prematur. Pada implementasi, dilakukan modifikasi dengan mengubah parameter-parameter dari susunan parameter artikel. Dari hasil percobaan, modifikasi susunan parameter menunjukkan solusi yang lebih baik.

---

Scheduling batching machines problem is a combinatorial problem to arrange jobs thus the processing time is minimum. Jobs have arbitrary size and processing time which will be assigned into batches. Total of jobs size in a batch can?t be more than machines capacity. Batch processing time is the longest jobs processing time inside the batch. While working, machines can?t be interrupted until all the jobs have done processed.

This problem is solved with improved ant colony optimization algorithm using metropolis criterion to prevent premature convergent solution. In implementation, parameter modification is made by changing parameters from the parameters arrangement of main article. Based on the result, the modification of parameters showed a better solution."