The layers of BP neural network model are input layer, hidden layer and output layer . The GM (1,1) model is based on the grey system theory. To give readers a clear understanding of the research methods used in this paper, the principles and applications of FA model, GM (1, 1) model and BP neural network are briefly introduced below.
The fifth part discusses the theoretical and practical contributions of this paper. In the national total express, e-commerce delivery accounts for a very high proportion reaching over 60%. However, some scholars (e.g. Hsiao et al. ; Kim et al. ) believe that based on the perspective of consumers, logistics systems can only play its role by integrating the whole e-commerce supply chain. For example, Inoue et al. showed that an ecosystem strategy based on the e-commerce market can significantly improve the performance of logistics companies. Demand forecasting equips leaders with actionable insights that go beyond the spreadsheet and gives them definitive information that clarifies choices and reduces uncertainty. The approach also provides better visibility into new market potential and demand variability, which enables smarter safety stock planning and tighter supply coordination.
Coupling analysis of logistics demand scale and principal component Through the coupling analysis of logistics demand scale with F1, F2 and F3 by curve fitting again, it can be found that logistics demand scale presents significant non-linear positive correlation with these three principal components. Therefore, factor analysis is used to reduce the dimension of these indicators. It can be found that the logistics demand scale has a significant non-linear positive correlation with the indicators of these three aspects. The curve fitting method is adopted to conduct a coupling analysis of the logistics demand scale with the commercial and trade environment, the basic supporting environment, and the e-commerce information environment.
External Factors
In this paper, a variety of influencing factors, such as commercial trade, basic support and e-commerce information, etc. are considered comprehensively. Since the data of each indicator are different in units, to get more accurate prediction results, this paper has carried out dimensionless processing on the original data of each indicator. The data for this paper were collected in August and September of 2020 when the Guangdong Statistical Yearbook for 2020 had not been published yet. The logistics demand scale is used to represent the logistics demand environment.
Therefore, this paper integrates the relevant indicators of e-commerce into the prediction of Guangdong logistics demand, reflects the scale of logistics demand more realistically, and expands the indicator system of logistics demand. For example, locker points of delivery (unattended) and service points of delivery (attended) . E-commerce logistics often involves both consumer and enterprise customers.
What are Logistics Forecasting Challenges?
For freight forwarders, demand forecasting predicts shipment volume by trade lane, mode, customer, and cargo type so contract, capacity, and staffing decisions can be made before the volume arrives. Demand forecasting in supply chain is the practice of predicting future demand for goods, shipments, or services based on historical patterns, current market conditions, and forward looking signals. In addition to the prediction of regional logistics demand, the implementation of new distribution modes by logistics enterprises is also helpful to alleviate problems such as insufficient capacity, and delay of logistics delivery during peak periods 13, 22. Therefore, it is necessary for relevant enterprises to further study the function and role of BP neural network model in logistics demand prediction.
Demand Forecasting Methods and Techniques in Supply Chain
It’s also measuring the supply chain cost of every business decision—and making sure those costs are considered from the start,” said the report. The insights gained from this analysis are then used to make informed decisions about inventory management, workforce planning, and other logistical aspects, ultimately leading to more efficient and cost-effective operations. This balance is crucial for minimizing holding costs and maximizing the availability of products for timely customer delivery. For freight forwarders, it predicts shipment volume by trade lane, mode, and customer so carrier contract, capacity, and staffing decisions can be made before the volume arrives instead of after. The importance of forecasting in supply chain, for a forwarder, is that every one of these decisions has a real cost when the plan is wrong.
Qualitative Forecasting Methods
The actual value, BP neural network model predicted value, and BP neural network prediction error value are shown in Table 8. Comparison of actual and BP neural network model predicted values It is found that the error between the BP neural network model predicted value and the actual value is small, indicating that the BP neural network prediction model has a high prediction accuracy.
Scientific and accurate prediction results can provide a reasonable reference for e-commerce platforms, and logistics enterprises to make decisions with the greatest effectiveness. First, relevant e-commerce platforms should pay attention to the prediction of regional logistics demand, especially in the peak period of logistics delivery such as a shopping carnival. The results show that GM (1, 1) model and BP neural network model have a good application prospect in regional logistics demand prediction, and BP neural network model has a relatively small prediction error and a relatively better prediction effect. Based on the historical background of e-commerce, this paper considers the related indicators of e-commerce driving logistics demand, and provides a new perspective for the establishment of regional logistics demand indicator https://thetimefinder.com/why-mobility-management-is-the-key-to-operational-agility/ systems. Since the 12 correlation degree values are all greater than 0.7 and reach the three-level accuracy , the 12 indicators selected in this study are applicable to the logistics demand scale prediction. According to the availability and authority of indicators, this paper selects the whole society’s cargo transport volume in Statistical Yearbook of Guangdong Province (2000–2019) to calculate the scale of regional logistics demand.
- E-commerce logistics often involves both consumer and enterprise customers.
- This balance is crucial for minimizing holding costs and maximizing the availability of products for timely customer delivery.
- This strategic approach not only streamlines operations but also supports better decision-making processes, ensuring that companies are better prepared to meet market demands and customer expectations.
- The demand forecasting process enhances forecasting accuracy in real-time, helps organizations manage their inventory levels and guides data-driven business decisions.
- From a practical point of view, the insights provided by our study can provide recommendations for logistics enterprises and relevant e-commerce platforms.
- Simulation methods are also used in logistics forecasting to model and assess the potential outcomes of different logistics scenarios.
Logistics Forecasting Methods
Coupling analysis of logistics demand https://scivast.com/articles/knowledge-based-systems-examples-applications/ scale and e-commerce information environment Through the coupling analysis of logistics demand scale and e-commerce information environment by curve fitting, it can be found that logistics demand scale is significantly positively correlated with e-commerce information environment. Coupling analysis of logistics demand scale and basic supporting environment Through the coupling analysis of logistics demand scale and foundation supporting environment by curve fitting, it can be found that logistics demand scale and foundation supporting environment also present a high positive correlation. Coupling analysis of logistics demand scale and commercial trade environment Through the coupling analysis of logistics demand scale and commercial trade environment by curve fitting, it can be found that logistics demand scale is significantly positively correlated with commercial trade environment.
- Modern forecasting combines quantitative methods, qualitative inputs from operators who see the market up close, and increasingly machine learning models that pick up patterns a human planner would miss.
- Ensuring data integrity involves not only collecting sufficient data but also regularly updating and verifying it to reflect current trends and patterns.
- Based on the historical background of e-commerce, this paper considers the related indicators of e-commerce driving logistics demand, and provides a new perspective for the establishment of regional logistics demand indicator systems.
- This reliability can help build trust and loyalty among customers, which is vital in today’s competitive market.
- Organizations seeking a minimally-invasive approach should consider passive demand forecasting.
Demand forecasting methods
When logistics operations are backed with AI, organizations benefit from enhanced forecast accuracy, improved decision-making, and increased efficiency. AI algorithms are capable of processing vast amounts of data, recognizing complex patterns, and learning from historical trends to make predictions about future logistics scenarios. This approach allows companies to test the effects of various strategies under simulated conditions before implementing them in the real world.