Bu kaydın yasal hükümlere uygun olmadığını düşünüyorsanız lütfen sayfa sonundaki Hata Bildir bağlantısını takip ederek bildirimde bulununuz. Kayıtlar ilgili üniversite yöneticileri tarafından eklenmektedir. Nadiren de olsa kayıtlarla ilgili hatalar oluşabilmektedir. MİTOS internet üzerindeki herhangi bir ödev sitesi değildir!

MULTI-PRODUCT, MULTI-STAGE PRODUCTION PLANNING MODEL AND DECISION SUPPORT SYSTEM SUGGESTION FOR F&B INDUSTRY

Oluşturulma Tarihi: 02-09-2016

Niteleme Bilgileri

Tür: Tez

Alt Tür: Doktora (Mühendislik)

Yayınlanma Durumu: Yayınlanmamış

Dosya Biçimi: PDF

Dil: İngilizce

Konu(lar): Üretim yönetimi. Operasyon yönetimi,

Yazar(lar): Tirkeş, Güzin (Yazar),

Emeği Geçen(ler): Çelebi, Neşe (Danışman),

Anahtar Kelimeler

Demand estimation in F&B industry, Time series, Trend analysis,Decomposition, Fuzzy, Holt-Winters, Optimization, Production planning, Mixedinteger linear programming, Decision support system 


Özet

A great deal of research has been undertaken in recent times on facility capacityexpansion and production planning problems under deterministic and stochasticdemand in the literature. However, only a small portion of this work directlyaddresses the issues faced by the food and beverage (F&B) industry especially insmall and medium size enterprises.In F&B industry; decision to increase the production capacity depends on a reliabledemand forecasting. In food production, forecasting the timing of demands is crucialin planning production scheduling to satisfy customer needs on time. Severalstatistical models have been used in demand forecasting in F&B industry. Most ofthe studies applied linear models such as autoregressive moving average (ARMA),autoregressive integrated moving average (ARIMA) for linear cases; non-linearARMA, Holt-Winters (HW) exponential smoothing, various kinds of data miningtechniques utilizing ensemble learning, artificial neural networks (ANN), geneticalgorithms with radial basis function and fuzzy logic systems used with non-lineardata, to predict future sales. Naturally, the model to be used for prediction is stronglyivrelated to the characteristics of the data such as the ‘trend’ or the ‘seasonality’observed. This study presents the results by examining the effectiveness of demandforecasting using time series analysis and possibilistic-probabilistic approaches inproduction scheduling problem of a real world multi-stage and multi-line sherbet andjam production company producing multiple products for both retail and wholesalesharing a limited capacity when demands are uncertain. A time series model for longterm forecasting is developed by obtaining monthly sales data from company fromJanuary 2013 to December 2014. Triple exponential smoothing method of HWmultiplicative with seasonality, trend analysis and seasonal decomposition methodsare compared and the best fit model is adapted in order to predict 2015 demand. Asperformance measures, the mean absolute percentage error (MAPE) ratio is used asevaluation metric. Following this approach possibilistic and probabilistic modelswere applied to predict the demand data for situations possessing uncertainties. Afterthe forecasting model, a mixed integer programming model is utilized for productionplanning and scheduling, containing a module caring for inventory planning. Theadaptability of the model to the cases dealing with ‘uncertainty’, keeps the model asan efficient base for the future studies with the given case. As the final stage, adecision support system was designed to help the users to deal with cases withvarying demand structures touching to extremities.


İçindekiler



Açıklamalar



Haklar



Notlar



Kaynakça


Atıf Yapanlar

Gözat Sayfasına Dön

 

Sosyal Medya ve Araçlar

İstatistikler

  • Kayıt
    • Bu ay: 0
    • Toplam: 2414
  • Online
    • Ziyaretçi: 16
    • Üye: 0
    • Toplam: 16

Detaylı İstatistikler