Production Planning Decision Based on the Optimized Fuzzy Time-series Clustering

Time series forecasting models based on a linear relationship model show great performance. However, these models cannot handle the the data that are incomplete, imprecise, and ambiguous as the interval-based fuzzy time series models since the process of fuzzification is abandoned. This article proposes a novel fuzzy time series forecasting model based on multiple linear regression and time series clustering for forecasting market prices. The proposed model employs a preprocessing to transform the set of fuzzy high-order time series into a set of high-order time series, with synthetic minority oversampling technique. After that, a high-order time series clustering algorithm based on the multiple linear regression model is proposed to cluster dataset of fuzzy time series and to build the linear regression model for each cluster. Then, we make forecasting by calculating the weighted sum of linear regression models’ results. Also, a learning algorithm is proposed to train the whole model, which applies artificial neural network to learn the weights of linear models. The interval-based fuzzification ensures the capability to deal with the uncertainties, and linear model and artificial neural network enable the proposed model to learn both of linear and nonlinear characteristics. The experiment results show that the proposed model improves the average forecasting accuracy rate and is more suitable for dealing with these uncertainties.

Download Full-text

Production Planning and Planting Pattern Scheduling Information System for Horticulture

E3S Web of Conferences ◽

10.1051/e3sconf/20183110004 ◽

2018 ◽

Vol 31 ◽

pp. 10004 ◽

Cited By ~ 1

Author(s):

Tanhella Zein Vitadiar ◽

Farikhin Farikhin ◽

Bayu Surarso

Keyword(s):

Time Series ◽

Information System ◽

Production Planning ◽

Research Result ◽

Fuzzy Time Series ◽

Series Method ◽

Planting Pattern ◽

Cropping Patterns ◽

Planning Decisions ◽

Time Series Method

This paper present the production of planning and planting pattern scheduling faced by horticulture farmer using two methods. Fuzzy time series method use to predict demand on based on sales amount, while linear programming is used to assist horticulture farmers in making production planning decisions and determining the schedule of cropping patterns in accordance with demand predictions of the fuzzy time series method, variable use in this paper is size of areas, production advantage, amount of seeds and age of the plants. This research result production planning and planting patterns scheduling information system with the output is recommendations planting schedule, harvest schedule and the number of seeds will be plant.

Download Full-text

Using Fuzzy Time Series (FTS) and Linear Programming for Production Planning and Planting Pattern Scheduling Red Onion

E3S Web of Conferences ◽

10.1051/e3sconf/201912523007 ◽

2019 ◽

Vol 125 ◽

pp. 23007 ◽

Cited By ~ 1

Author(s):

Aries Dwi Indriyanti ◽

Dedy Rahman Prehanto ◽

Ginanjar Setyo Permadi ◽

Chamdan Mashuri ◽

Tanhella Zein Vitadiar

Keyword(s):

Time Series ◽

Linear Programming ◽

Production Planning ◽

Fuzzy Time Series ◽

Planning System ◽

Cropping Pattern ◽

Market Demand ◽

Time Data ◽

Cropping Patterns ◽

Data Production

This study discusses the production planning system and scheduling shallots planting patterns using fuzzy time series and linear programming methods. In this study fuzzy time series to predict the number of requests and the results of predictions from fuzzy time series methods become one of the variables in the calculation of linear programming. Using supporting variables, demand data, production data, employment data, land area data, production profit data, data on the number of seedlings and planting time data are indicators used in processing the system. The system provides recommendations for cropping patterns and the number of seeds that must be planted in one period. The age of harvesting onions is ± 3-4 months from the planting process, the number of planting seeds is adjusted to the number of requests that have been predicted by using fuzzy time series and cropping pattern calculation process is carried out using linear programming. The results of this system are recommendations for farmers to plant seedlings, planting schedules, and harvest schedules to meet market demand.

Download Full-text

Application of Fuzzy Time Series (FTS) Algorithm in Production Planning of Indonesia’s Oil Refining Company

Indonesian Scholars Scientific Summit Taiwan Proceeding ◽

10.52162/3.2021108 ◽

2021 ◽

Vol 3 ◽

pp. 16-23

Author(s):

Zakka Ugih Rizqi ◽

Tommy Aries Kurniawan ◽

Adinda Khairunisa

Keyword(s):

Time Series ◽

Production Planning ◽

Production Cost ◽

Fuzzy Time Series ◽

Oil Refining ◽

Hybrid Strategy ◽

Starting Point ◽

Resource Requirements ◽

Production Period ◽

Production Target

Forecasting and aggregate planning are crucial phases in production planning especially for oil refining company that takes expensive production cost. Accurate forecasting greatly influences the success of production planning since it is the starting point of production planning. Whereas aggregate planning becomes important because it functions to bridge between the demand or production target with the existing resource requirements. Seeing the importance of accurate forecasting and aggregate planning, this research emphasizes the use of Fuzzy Time Series (FTS) Algorithm to forecast Premium sales in Indonesia’s oil refining company. The comparison is also done between FTS with the other classical techniques in time series forecasting to test the reliability of algorithm and FTS outperforms the others based on the lowest MAPE value as much as 0.87%. FTS result is then used as an input in the aggregate planning by using heuristics method and comparing 3 strategies which are Level Strategy, Chase Strategy, and Hybrid Strategy. The result shows that Hybrid Strategy is the most efficient one because it produces the lowest production cost for three months production period as much as Rp 3,272,000,000.

Download Full-text

About One Approach to Defuzzifikation of Outputs of Fuzzy Time Series Model

Indian Journal Of Applied Research ◽

10.15373/2249555x/sept2013/173 ◽

2011 ◽

Vol 3 (9) ◽

pp. 562-566

Author(s):

Ramin Rzayev ◽

◽

Musa Agamaliyev ◽

Nijat Askerov

Keyword(s):

Time Series ◽

Time Series Model ◽

Fuzzy Time Series

Download Full-text

Peramalan Data IHSG Menggunakan Metode Backpropagation

Jurnal ULTIMATICS ◽

10.31937/ti.v5i1.310 ◽

2013 ◽

Vol 5 (1) ◽

pp. 26-30

Author(s):

Seng Hansun

Keyword(s):

Time Series ◽

Soft Computing ◽

Fuzzy Time Series

Jaringan saraf tiruan merupakan salah satu metode soft computing yang banyak digunakan dan diterapkan di berbagai disiplin ilmu, termasuk analisis data runtun waktu. Tujuan utama dari analisis data runtun waktu adalah untuk memprediksi data runtun waktu yang dapat digunakan secara luas dalam berbagai data runtun waktu real, termasuk data harga saham. Banyak peneliti yang telah berkontribusi dalam analisis data runtun waktu dengan menggunakan berbagai pendekatan berbeda. Chen dan Hsu, Jilani dkk., Stevenson dan Porter, dan Hansun telah menggunakan metode fuzzy time series untuk meramalkan data mendatang, sementara beberapa peneliti lainnya menggunakan metode hibrid, seperti yang dilakukan oleh Subanar dan Suhartono, Popoola dkk, Popoola, Hansun dan Subanar. Di dalam penelitian ini, penulis mencoba untuk menerapkan metode jaringan saraf tiruan backpropagation pada salah satu indikator perubahan harga saham, yakni IHSG (Indeks Harga Saham Gabungan). Penelitian dilanjutkan dengan menghitung tingkat akurasi dan kehandalan metode yang telah diterapkan pada data IHSG. Pendekatan ini diharapkan dapat menjadi salah satu cara alternatif dalam meramalkan data IHSG sebagai salah satu indikator perubahan harga saham di Indonesia. Kata kunci—jaringan saraf tiruan, backpropagation, analisis data runtun waktu, soft computing, IHSG

Download Full-text