Artificial intelligence-based evaluation of the factors affecting the sales of an iron and steel company

Mehmet Pekkaya · Zafer Uysal · Aytaç Altan · Seçkin Karasu

38 citations0/0 files storedTURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES

No readable copy stored yet — this record is metadata only.

No open-access location is known for this record yet.

Literature note Cite in a project Citation map

Forecasting Techniques and Applications Neural Networks and Applications Stock Market Forecasting Methods

Abstract

It is important to predict the sales of an iron and steel company and to identify the variables that influence these sales for future planning. The aim in this study was to identify and model the key factors that influence the sales volume of an iron and steel company using artificial neural networks (ANNs). We attempted to obtain an integrated result from the performance/sales levels of 5 models, to use the ANN approach with hybrid algorithms, and also to present an exemplary application in the base metals industry, where there is a limited number of studies. This study contributes to the literature as the first application of artificial intelligence methods in the iron and steel industry. The ANN models incorporated 6 macroeconomic variables and price-to-sales data and their results were evaluated. An ordinary least squares regression model was also used to facilitate the comparison of results, while gray relational analysis (GRA) was used to draw a comprehensive conclusion based on the ANN results. The results showed that the variables USD/TL exchange rate, product prices, and interest rates, in descending order, had the highest degree of influence in determining the sales of the iron and steel company. Furthermore, these variables are crucial for forecasting future sales and strategic planning. The study showed that the ANN outperformed classical regression models in terms of prediction accuracy. In the model applications conducted for 5 different product groups, it was observed that 3 models (models 2, 3, and 4), including model 4, which sold a higher volume of products than the total of the other products, had an overall performance above 80%. In addition, GRA was found to be a valuable tool for synthesizing insights from different ANN models based on their respective performance levels.

Paper brief

Loading…

Cite this work

Citation unavailable for this work.

BibTeX RIS CSL-JSON

No public copy is stored for this record, so only the metadata is cited.

Citations in this library

Citation map →

Stored only when both papers are in the corpus — 38 citations exist upstream.

References

2020A new forecasting model with wrapper-based feature selection approach using multi-objective optimization technique for chaotic crude oil time series 2019Digital currency forecasting with chaotic meta-heuristic bio-inspired signal processing techniques 2008A decision support system for demand forecasting with artificial neural networks and neuro-fuzzy models: A comparative analysis 2011Determining of stock investments with grey relational analysis 2005Evolving neural network for printed circuit board sales forecasting 2013An alternative approach to estimating the parameters of a generalised Grey Verhulst model: An application to steel intensity of use in the UK 2017Pattern graph tracking-based stock price prediction using big data 2010Performance analysis of demand planning approaches for aggregating, forecasting and disaggregating interrelated demands

Cited by

2024Exploring climate change discourse on social media and blogs using a topic modeling analysis 2024Fuzzy Integrated Delphi-ISM-MICMAC Hybrid Multi-Criteria Approach to Optimize the Artificial Intelligence (AI) Factors Influencing Cost Management in Civil Engineering

Explore the map →

Papers that cite the same references — the relational fallback while embeddings are still backfilling.

Imported from OpenAlex (CC0) — nothing here is generated.