Table 4 - uploaded by Michal Paták
Content may be subject to copyright.
Context in source publication
Similar publications
Citations
... A great share of such costs arises either from excessive or insufficient coverage. Since demand forecasting plays a vital role in the control of retailers' supply chains, an improvement of its accuracy can contribute to significant cost reduction [1,2,3,4]. ...
In this paper we investigate to what extent long short-term memory neural networks (LSTMs) are suitable for demand forecasting in the e-grocery retail sector. For this purpose, univariate as well as multivariate LSTM-based models were developed and tested for 100 fast-moving consumer goods in the context of a master's thesis. On average, the developed models showed better results for food products than the comparative models from both statistical and machine learning families. Solely in the area of beverages random forest and linear regression achieved slightly better results. This outcome suggests that LSTMs can be used for demand forecasting at product level. The performance of the models presented here goes beyond the current state of research, as can be seen from the evaluations based on a data set that unfortunately has not been publicly available to date.
... Demand forecasting plays a key role in the management of logistics processes in retailing. Therefore, this quantitative research deals with the demand forecasting of retail store in Czech republic (Patak et al., 2015). The research methodology works on 75 retail stores which has a section of groceries, data collection took place in the period of March-April 2015 using face-to-face interviews. ...
Forecasting is an essential component of good financial management and informed
decision-making. Effective forecasting requires organizations to recognize that forecasts are more
than a technical activity, and emphasize their importance to financial and operational management.
It is essential that departments generate cooperation and understanding between the analysts who
produce forecasts, and their policy, operational and finance colleagues who use them to manage
the business.
The aim of this research is to forecast the future needs of demands for Ministry of Education in
order to adjust the suitable strategic and financial plan due to the lack of using the information
technology tools which is the main problem we try to solve.
In this research, we collect our data from Ministry of Education in Gaza Strip database. From the
collected data we extract the useful features that help our achievements, which is to forecast the
needed demands in medium term for the next five years. The data extracted consists of the voucher
date, which is the transaction date combined into monthly record, and quantity which represent
the amount of item transaction in this month. We apply five different methods of forecasting on
the dataset. The methods are Simple Average, Moving Average, Holt-Winter, Auto-Regressive
Integrated Moving Average (ARIMA) and Hybrid ARIMA combined with Neural Network (NN).
The performance after applying these methods shows that Hybrid ARIMA method has the best
performance with lowest error rate. The results of applying Hybrid ARIMA on the sample dataset
according to Mean Absolute Percentage Error (MAPE) shows that for the original data is 17.65%
and for the modified data with replacing null values with average it give 16.31%.
The results shows after forecasting for the next five years from 2020 to 2024 and comparing the
results with the period from 2015 to 2019 that amount needs of Tables will increase by 3% and
Chairs will increase by 47%. The selected demands items such as A4 papers will increase by 9%,
white and colored Bristol papers will decrease by 3% and 24%, white and colored chalks will
increase by 11%, pens will decrease by 15%, pencils will decrease by 38% and marker pens will
decrease by 4%.
Based on the results, we expect an increasing of some demands needing, and decreasing for the
others in the next five years.
