Refine
Document Type
- Doctoral Thesis (9) (show_all)
Language
- English (9) (show_all)
Has Fulltext
- yes (9)
Is part of the Bibliography
- no (9)
Keywords
- Supply Chain Management (9) (show_all)
This dissertation investigates the value of customer behavior in supply chain management through the application of (big) data analytics in demand forecasting. The use of advanced analytics in supply chain management is not novel. However, the growing expansion of data volumes provides companies with new opportunities to optimize their supply chain. Despite the rising interest from both academia and practice and the recent increase in publications in this area, empirical insights are still limited. At the same time, changes in customer expectations towards instant product delivery require companies to rethink their supply chain, where accurate demand forecasts are often at the core of enabling efficient and flexible processes. This makes the development of demand prediction models that can be used in practice especially relevant. We1 analyze the use of customer behavior in demand forecasting in three separate research papers. Leveraging data from research partners in the online fashion and construction industry, we assess the potential of the developed prediction models in three areas of application in supply chain management, namely order fulfillment, order picking, and inventory planning. In the first paper, we develop a prediction model for anticipatory shipping in the fashion industry, which predicts customers’ online purchases with the aim of shipping products in advance, and subsequently minimizing delivery times. Using various forecasting methods and data on customers’ behavior on the website, we test if, and how early, it is possible to predict online purchases. Results indicate that customer purchases are, to a certain extent, predictable, but anticipatory shipping comes at a high cost due to wrongly sent products. The second paper assesses the extent to which clickstream data can improve forecast accuracy for fashion products. Specifically, we assess which clickstream variables are most suitable for predicting demand, and identify the products that benefit most from this. Results indicate that clickstream data is especially useful for forecasting medium- and certain intermittent-demand products. A simulation of order picking for these products shows that using clickstream data in the forecast substantially decreases picking times. The third paper investigates how sequential pattern mining can be used to determine products with correlated demand, and how to leverage this as an input into forecasting for a supplier in the construction industry. We find that sequential pattern mining may be beneficial when used in combination with traditional forecasting methods, and that support vector regression models seem especially suited to forecast intermittent-demand products. An application to inventory planning shows that our developed forecasting model might reduce the company’s costs of inventory holding and lost sales by up to 6.9%. Overall, our research highlights the value of using customer behavior to enhance demand forecasting and the benefit of using improved forecasts in various applications in supply chain management.
1Referring to the authors of the respective chapters as noted at the beginning of each chapter.
Supply chain risk management
(2012)
Supply chain fit
(2009)
This dissertation investigates different applications of data analytics in supply chain planning. In the last years, data analytics became more important, because of the increase of computational power and the larger availability of data. Data analytics is used in various domains to improve operations performance, increase customer satisfaction and revenues. However, both the research and the application of data analytics in supply chain management is still lacking behind other industries. We1 analyze the potential of data analytics in the field of supply chain planning in three exemplary fields: demand forecasting, partial defection prediction and price discrimination. In addition, we demonstrate how to deal with three common challenges in the field of data analytics: the manual effort for method selection and hyperparameter tuning, the difficult interpretability of machine learning methods and the risks associated with data collection through randomized experiments. In the first paper, we develop a method selection approach in the field of intermittent demand prediction. Our model combines high predictive performance with automation and calculation efficiency. Unlike common practice, the prediction method gets automatically chosen for each data set without any manual selection. Our results are stable across three different data sets that come from different sources but all contain intermittent demand time series. We showcase the impact of the proposed forecasting approach with a warehouse operation simulation. We thereby prove the financial benefit with empirical data. In the second paper, we deal with partial defection prediction in a business-tobusiness environment in the logistics industry. The predictions must combine predictive performance with interpretability and profit maximization. Our model uses a large variety of customer-based and time-series-based features to predict the probability of partial defection for each customer. We use a data permutation approach to make the best performing, black-box models interpretable. Furthermore, we use a profit assessment to identify the method that leads to the highest revenue through successful retention actions. In the third paper, we study price sensitivity prediction. We do not use any randomized experiments, because the risk of loosing customers through such experiments is too high. Thereby, we address the challenge of data availability
1The term “we” refers to the authors of the respective chapters as denoted at the beginning of each chapter. For the abstract, this refers to the authors of Faber and Spinler (2019a,b,c).