International Journal of Value Engineering

Enhancing procurement demand forecasting using advanced artificial intelligence techniques in medium-sized enterprises

Dr. AtmaNand

School of Applied and Life Sciences, Uttaranchal University

Shiva Nand

Department of Phsarmacy, Shyam Sai Institute of Education

Keywords: artificial intelligence; deep learning; forecasting; procurement; time series prediction; medium-sized enterprises; LSTM; attention mechanism

---

Abstract

This paper examines the application of advanced artificial intelligence (AI) methods to optimize demand forecasting in procurement within medium-sized enterprises. Building upon previous work that demonstrated the superiority of basic artificial neural networks (ANNs) over traditional statistical methods, this research integrates state-of-the-art deep learning techniques, including long short-term memory (LSTM) networks and attention mechanisms, to further enhance forecasting accuracy. We provide a comprehensive literature survey, represented graphically, to contextualize our approach within current research trends. Mathematical formulations of the models and solutions are presented to detail the methodologies used. We compare these advanced AI models with traditional methods such as ARIMA and baseline approaches within the empirical context of the REIFF Group, a prominent medium-sized B2B retailer. Our findings, supported by graphs and diagrams, indicate that the incorporation of more sophisticated AI techniques significantly improves forecast precision, particularly in handling non-linear and complex time series data. Challenges specific to medium-sized enterprises, such as data quality, resource constraints, and the need for explainability, are addressed, and strategies to mitigate these challenges are discussed. This work contributes to the field by demonstrating the practical viability of advanced AI forecasting models in the procurement processes of medium-sized enterprises and offers insights into their integration within existing business practices.

---

References

[1] Asadi, S. and Omran, M. G. (2020). Demand forecasting using machine learning algorithms. Journal of Industrial Engineering International , 16(1):1–6.

[2] Bahdanau, D., Cho, K., and Bengio, Y. (2015). Neural machine translation by jointly learning to align and translate. ICLR 2015 .

[3] Bengio, Y., Simard, P., and Frasconi, P. (1994). Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks, 5(2):157–166.

[4] Hochreiter, S. and Schmidhuber, J. (1997). Long short-term memory. Neural Computation , 9(8):1735–1780.

[5] Hyndman, R. J. and Athanasopoulos, G. (2018). Forecasting: Principles and Practice . OTexts.

[6] Kumar, V. (2020). Application of mathematical modeling in population forecasting. Uttaranchal University Journal of Applied and Life Sciences, 3(1):50–57.

[7] Lea, C., Flynn, M. D., Vidal, R., Reiter, A., and Hager, G. D. (2017). Temporal convolutional net- works for action segmentation and detection. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,pages 156–165.

[8] Nand, A. (2023a). Deteriorating inventory management model: adapting to variable lead time and consumer-driven shortages. International Journal of Operational Research . DOI: 10. 1504/IJOR.2023.10064634.

[9] Nand, A. (2023b). Dynamic terminology for supplier–retailer collaboration metrics in inven- tory model. International Journal of Mathematics in Operational Research . DOI: 10.1504/IJ- MOR.2024.10064476.

[10] Nand, A. (2024a). Linear and non-linear time dependent deteriorated inventory control model. Journal of Advanced Manufacturing Systems. DOI: 10.1142/S0219686725500039.

[11] Nand, A. (2024b). Optimized economic production quantity inventory model with Weibull decay, dynamic holding cost under varied demand scenario and demand series analysis. Journal of Advanced Manufacturing Systems. DOI: 10.1142/S0219686725500106.

[12] Nand, A. (2024c). Next-generation inventory optimization: advanced inventory management har- nessing demand variability integrating fuzzy logic and granular differentiability. RAIRO - Op- erations Research. DOI: 10. 1051/ro/2024226.

[13] Nand, A. and Chauhan, N. S. (2024). Multi-entity stock dependent model with capacity and manufacture cost restraints. Malaysian Journal of Science , 43:37–48.

[14] Nand, A., Shivanand, V. K., Chauhan, N. S., and Kumar, V. (2022). Inventory control model for pharmaceutical items with exponential demand and holding cost varying based on time. International Journal of Mechanical Engineering , 7:550–557.

[15] Srivastava, V. K. and Nand, A. (2023). Application and technique of inventory control theory in pharmaceutical sciences. International Journal of Operational Research , 47:109–123.