ePrints.FRI - University of Ljubljana, Faculty of Computer and Information Science

Demand forecasting

Belčec Gregor (2011) Demand forecasting. EngD thesis.

[img] PDF
Download (1831Kb)


    Companies operate in an increasingly challenging environment that requires them to continuously improve all areas of the business process. Demand forecasting is one area in manufacturing companies where we can hope to gain great advantages. Improvements in forecasting can result in cost savings throughout the supply chain, improve the reliability of information and the quality of the service for our customers. In the company Danfoss Trata, d. o. o. we did not have a system for demand forecasting. The forecast was made on the basis of historical data and occasional information from the market. The use of a demand forecasting system was thus an urgently needed improvement. Despite the various solutions found on the market today we have decided to create our own forecasting system. The cost of development and implementation, and user involvement in the development and upgrade of process were two important advantages of this. Users have thus gained a deeper understanding of the factors affecting the forecast. The biggest challenge was creating our own forecasting algorithm, based on the specific business environment in which the company Danfoss Trata, d. o. o. operates. Thus, the first part of the diploma paper includes an introduction to the field of demand forecasting as well as an explanation on why forecasting is so important. It includes a few guidelines for selecting the most appropriate forecasting model and introduces time series, the understanding of which has played an important role when developing our own forecasting algorithm. This is followed by an introduction of the new forecasting process of the company Danfoss Trata, d. o. o. and the value of forecasts in various circumstances. In the end, the forecasting algorithm, which was later implemented in the Danfoss demand forecasting system, is presented in detail.

    Item Type: Thesis (EngD thesis)
    Keywords: Demand forecasting, time series, statistical models, supply chain management, demand planning.
    Number of Pages: 65
    Language of Content: Slovenian
    Mentor / Comentors:
    Name and SurnameIDFunction
    prof. dr. Marko Bajec245Mentor
    Link to COBISS: http://www.cobiss.si/scripts/cobiss?command=search&base=50070&select=(ID=00008478036)
    Institution: University of Ljubljana
    Department: Faculty of Computer and Information Science
    Item ID: 1407
    Date Deposited: 27 Jun 2011 08:58
    Last Modified: 13 Aug 2011 00:39
    URI: http://eprints.fri.uni-lj.si/id/eprint/1407

    Actions (login required)

    View Item