Experiments on Combining Demand Forecasts with Semiconductor Data

نویسندگان

  • Feng Zhang
  • Aliza Heching
  • Sarah Hood
  • Jonathan Hosking
  • Ying-tat Leung
  • Robin Roundy
  • Johnathan Wong
چکیده

In this paper we test several forecast combination methods on industrial demand and forecast data from a semiconductor company. The tested combination methods include four popular methods existing in the literature, and new methods proposed in a companion paper. There are two forecasts to be combined: a time series forecast and a marketing forecast. According to the experiments, combining different demand forecasts does help to improve accuracy. The new combination methods proposed by authors have better and more stable performance than the others. The application of appropriate non-linear transformations to the original demand and forecast data before feeding it into the combination models is strongly recommended.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A Model for Master Planning in Semiconductor Manufacturing

We present a model for making master planning decisions in semiconductor manufacturing. The problem consists in determining appropriate wafer quantities for certain products, facilities (locations), and periods of time. Different demand types, i.e., confirmed orders or forecasts are considered. The reentrant flow in semiconductor wafer fabrication facilities is taken into account for capacity m...

متن کامل

Modelling and Forecasting Australian Domestic Tourism

In this paper, we model and forecast Australian domestic tourism demand. We use a regression framework to estimate important economic relationships for domestic tourism demand. We also identify the impact of world events such as the 2000 Sydney Olympics and the 2002 Bali bombings on Australian domestic tourism. To explore the time series nature of the data, we use innovation state space models ...

متن کامل

Using Combined Forecasts with Changing Weights for Electricity Demand Profiling

Day-ahead half-hourly demand forecasts are required for scheduling and for calculating the daily electricity pool price. One approach predicts turning points on the demand curve and then produces half-hourly forecasts by a heuristic procedure, called profiling, which is based on a past demand curve. This paper investigates possible profiling improvements. Using a cubic smoothing spline in the h...

متن کامل

Freight transport development forecasts in enterprises management

In this study, an investigation of long-term forecasts relating to the development of the transport sector in Poland is performed, including the ones by 2030 and 2050. Selected transport development forecasts from the perspective of the membership of Poland in the European Union are presented and most of all, from the perspective of national studies. The basement for the review was the prognosi...

متن کامل

A collaborative demand forecasting process with event-based fuzzy judgments

Mathematical forecasting approaches can lead to reliable demand forecast in some environments by extrapolating regular patterns in time-series. However, unpredictable events that do not appear in historical data can reduce the usefulness of mathematical forecasts for demand planning purposes. Since forecasters have partial knowledge of the context and of future events, grouping and structuring ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2004