Supply Chain Forecasting

Introduction:

Maintaining effective supply chain operations is crucial for firms to deliver finished goods to clients as market competition rises. When retailers send order quantities to their suppliers in response to demand, they must carefully balance inventory levels with anticipated future demand.

In today's increasingly dynamic world, where a variety of natural and man-made disastrous events, such as pandemics, earthquakes, equipment failures, labor disputes, supplier defaults, and political instability affect the performance of operations and frequently result in financial damage, the supply chain plays a very important role. The creation of resilient supply chain systems is one method for addressing the high levels of global supply chain risks and market disruptions that stop the flow of goods and information between raw materials, industry, and the end customer. A supply chain must be able to forecast effectively and predict disruptions, in order to lessen their effects, should they occur or completely avoid them. It must also be able to promptly react to and recover from these costly disruptions.

The bullwhip effect, which is described as the upstream amplification of demand fluctuation, causes supply chains to have issues guaranteeing that the inventory can satisfy customer demand. This topic has drawn a lot of attention from researchers in the field of supply chain management due to the wide range of its effects and its prevalence in practice. It is also crucial for practitioners who want to meet their performance goals and enhance their operations at both the local and global levels of the supply chain. Demand forecasting in a supply chain, where there are certain difficulties, is one such invention to enhance operational effectiveness. Higher demand and inaccurate forecasts cause more inventory to be retained, which negatively affects inventory turnover and the company's profitability. Poor predictions cause a business to keep insufficient safety supplies, which raises inventory costs, lowers service levels, and increases order volatility. Therefore, a forecasting system was developed to be integrated with our existing supplier relationship management product JetSRM Supplier Portal, which is a platform for data integration that facilitates information exchange between suppliers and enterprises, utilizing statistical methods that are based on historical data and AI techniques that use time series data to estimate future developments. In JetSRM, the data gathered from the suppliers is sent to the corporate systems, and the associated procedures are documented in the SAP system. Processes are tracked through a single interface, eliminating transactions carried out by fax, phone, and email. The portal assists the optimization of the procurement processes and maintains the progressiveness of corporate and supplier information. The transfer of potential and current suppliers to the system is the first step in the portal usage process, which continues with the offer procedures, sample requests, and order requests.

Methodology:

Due to the data having seasonality and trend significance, our system adopts both statistical methods such as ARIMA (Autoregressive Integrated Moving Average), SARIMA (Seasonal Autoregressive Integrated Moving Average),  and artificial intelligence-based methods such as XGBoost (eXtreme Gradient Boosting) that often put the above-mentioned significance into consideration during forecasting. The effectiveness of the models built using the provided data is directly affected by the choice of forecasting methodologies. Thus, after the development of the models, forecasting and testing is conducted to evaluate the most suitable model to effectively lower inventory costs and increase overall profitability.

Conclusions:

Accurate forecasts obtained from models with the highest levels of precision guarantee a flexible and agile supply chain. They also enable businesses to respond to disruptions and eliminate several inefficiencies like higher costs associated with producing more than is necessary, wastage, and transportation costs. Reducing the cost of supply chain members results in lower product prices for customers, which is one of the main concerns in competitive business environments.

[1] Nagaraja, C. H., Thavaneswaran, A., and Appadoo, S. S., “Measuring the bullwhip effect for supply chains with seasonal demand components,” European Journal of Operational Research, vol. 242, pp. 445–454, 2015.

[2] Steven A. M, Closs, D. J., Griffis, S. E., Zobel, C. W., and Macdonald, J. R., https://www.supplychain247.com/article/understanding supply chain resilience. November, 2015.