Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
241
result(s) for
"Inventories, Retail Data processing."
Sort by:
The impact of order fulfillment on consumer experience: text mining consumer reviews from Amazon US
PurposeThis research analyzes online consumer reviews and ratings to assess e-retail order fulfillment performance. The study aims to (1) identify consumer journey touchpoints in the order fulfillment process and (2) determine their relative importance for the consumer experience.Design/methodology/approachText mining and analytics were employed to examine over 100 m online purchase orders, along with associated consumer reviews and ratings from Amazon US. Using natural language processing techniques, the corpus of reviews was structured to pinpoint touchpoints related to order fulfillment. Reviews were then classified according to their stance (either positive or negative) toward these touchpoints. Finally, the classes were correlated with consumer rating, measured by the number of stars, to determine the relative importance of each touchpoint.FindingsThe study reveals 12 touchpoints within the order fulfillment process, which are split into three groups: delivery, packaging and returns. These touchpoints significantly influence star ratings: positive experiences elevate them, while negative ones reduce them. The findings provide a quantifiable measure of these effects, articulated in terms of star ratings, which directly reflect the influence of experiences on consumer evaluations.Research limitations/implicationsThe dataset utilized in this study is from the US market, which limits the generalizability of the findings to other markets. Moreover, the novel methodology used to map and quantify customer journey touchpoints requires further refinement.Practical implicationsIn e-retail and logistics, comprehending touchpoints in the order fulfillment process is pivotal. This understanding helps improve consumer interactions and enhance satisfaction. Such insights not only drive higher conversion rates but also guide informed managerial decisions, particularly in service development.Originality/valueDrawing upon consumer-generated data, this research identifies a cohesive set of touchpoints within the order fulfillment process and quantitatively evaluates their influence on consumer experience using star ratings as a metric.
Journal Article
Transformer-Based Models for Probabilistic Time Series Forecasting with Explanatory Variables
Accurate demand forecasting is essential for retail operations as it directly impacts supply chain efficiency, inventory management, and financial performance. However, forecasting retail time series presents significant challenges due to their irregular patterns, hierarchical structures, and strong dependence on external factors such as promotions, pricing strategies, and socio-economic conditions. This study evaluates the effectiveness of Transformer-based architectures, specifically Vanilla Transformer, Informer, Autoformer, ETSformer, NSTransformer, and Reformer, for probabilistic time series forecasting in retail. A key focus is the integration of explanatory variables, such as calendar-related indicators, selling prices, and socio-economic factors, which play a crucial role in capturing demand fluctuations. This study assesses how incorporating these variables enhances forecast accuracy, addressing a research gap in the comprehensive evaluation of explanatory variables within multiple Transformer-based models. Empirical results, based on the M5 dataset, show that incorporating explanatory variables generally improves forecasting performance. Models leveraging these variables achieve up to 12.4% reduction in Normalized Root Mean Squared Error (NRMSE) and 2.9% improvement in Mean Absolute Scaled Error (MASE) compared to models that rely solely on past sales. Furthermore, probabilistic forecasting enhances decision making by quantifying uncertainty, providing more reliable demand predictions for risk management. These findings underscore the effectiveness of Transformer-based models in retail forecasting and emphasize the importance of integrating domain-specific explanatory variables to achieve more accurate, context-aware predictions in dynamic retail environments.
Journal Article
Goodbye Pareto Principle, Hello Long Tail: The Effect of Search Costs on the Concentration of Product Sales
Many markets have historically been dominated by a small number of best-selling products. The Pareto principle, also known as the 80/20 rule, describes this common pattern of sales concentration. However, information technology in general and Internet markets in particular have the potential to substantially increase the collective share of niche products, thereby creating a longer tail in the distribution of sales. This paper investigates the Internet's \"long tail\" phenomenon. By analyzing data collected from a multichannel retailer, it provides empirical evidence that the Internet channel exhibits a significantly less concentrated sales distribution when compared with traditional channels. Previous explanations for this result have focused on differences in product availability between channels. However, we demonstrate that the result survives even when the Internet and traditional channels share exactly the same product availability and prices. Instead, we find that consumers' usage of Internet search and discovery tools, such as recommendation engines, are associated with an increase the share of niche products. We conclude that the Internet's long tail is not solely due to the increase in product selection but may also partly reflect lower search costs on the Internet. If the relationships we uncover persist, the underlying trends in technology portend an ongoing shift in the distribution of product sales.
This paper was accepted by Ramayya Krishnan, information systems.
Journal Article
Bullwhip Effect Measurement and Its Implications
The bullwhip effect, or demand information distortion, has been a subject of both theoretical and empirical studies in the operations management literature. In this paper, we develop a simple set of formulas that describe the traditional bullwhip measure as a combined outcome of several important drivers, such as finite capacity, batch-ordering, and seasonality. Our modeling framework is descriptive in nature as it features certain plausible approximations that are commonly employed in practical inventory systems. The results are nonetheless compelling and can be used to explain various conflicting observations in previous empirical studies. Building on the theoretical framework, we discuss the managerial implications of the bullwhip measurement. We show that the measurement can be completely noninformative about the underlying supply chain cost performance if it is not linked to the operational details (such as decision intervals and leadtimes). Specifically, we show that an aggregated measurement over relatively long time periods can mask the operational-level bullwhip. In addition, we show that masking also exists under product or location aggregation in some illustrative cases.
Journal Article
A transformer-based framework for enterprise sales forecasting
Sales forecasting plays an important role in business operations as it impacts decisions on inventory management, allocation of resources, and financial planning. Accurate sales predictions are essential for optimizing cash flow management, adapting marketing and sales strategies, and facilitating strategic planning. This study presents a computational framework for predicting business sales using transformers, which are considered one of the most powerful deep learning architectures. The design of our model is specifically tailored to accommodate tabular data with low dimensions. The experimental results demonstrated that our proposed method surpasses conventional machine learning models, achieving reduced mean absolute error (MAE), mean square error (MSE), and root mean square error (RMSE), as well as higher R 2 values of nearly 0.95. The results confirmed that the model is applicable not only to this research but also to similar studies that use low-dimensional tabular data. The improved accuracy and stability of our model demonstrate its potential as a useful tool for enhancing sales prediction, therefore facilitating more informed decision-making and strategic planning in corporate operations.
Journal Article
An Empirical Examination of the Decision to Invest in Fulfillment Capabilities: A Study of Internet Retailers
Internet technology has allowed for a higher degree of decoupling between the information-intensive sales process and the physical process of inventory management than its brick-and-mortar counterpart. As a result, some Internet retailers choose to outsource inventory and back-end operations to focus on the sales/marketing aspects of e-commerce. Nonetheless, many retailers keep fulfillment capabilities in-house. In this paper, we identify and empirically test factors that persuade firms to integrate inventory and fulfillment capabilities with virtual storefronts. Based on the extant literature and previous research in e-commerce, we formulate nine theoretical predictions. We then use data from a sample of over 50 public Internet retailers to test whether empirical data are consistent with these hypotheses. Finally, given the strategic importance and financial magnitude of the inventory investment decision, we analyze the effect of this decision on the economic success of Internet retailers during the period of study. We find that there are many circumstances in which it is prudent to own fulfillment capabilities and inventory. Empirical data are consistent with hypotheses that this tendency is higher for older firms selling small, high-margin products, offering lower levels of product variety, and facing lower demand uncertainty. We also discover that firms making inventory ownership decisions that are consistent with an empirical benchmark derived from environmental and strategic factors are less likely to go bankrupt than those making inconsistent inventory choices.
Journal Article
Predicting Trends and Maximizing Sales: AI’s Role in Saudi E-Commerce Decision-Making
Artificial intelligence (AI) has emerged as a transformative force across various sectors, providing innovative solutions and enhancing operational processes. In the e-commerce domain, AI has significantly contributed to customer-centric approaches and strategic decision-making, fostering superior customer experiences. This study investigates the role and impact of AI in the Saudi e-commerce sector, focusing on the perspectives of female customers and retailers. Grounded in sociotechnical theory, the research employs a mixed-methods approach, combining quantitative surveys and semi-structured interviews. The quantitative findings demonstrate that AI-enabled e-commerce positively influences customer experience, customer satisfaction, and operational efficiency. Key AI capabilities, such as task automation, personalized recommendations, and predictive analytics, enhance online retail systems’ performance. The qualitative analysis highlights both the opportunities and challenges associated with AI adoption, emphasizing the need for specialized infrastructure and skilled professionals. Participants recommend addressing the skill gap and adopting phased implementation strategies to optimize AI integration. This study provides actionable insights and strategic recommendations for policymakers and stakeholders in the Saudi e-commerce sector.
Journal Article
Leveraging Market Basket Analysis for Targeted Promotions and Personalized Recommendations
This project develops a comprehensive web application that integrates advanced machine learning techniques with traditional retail analysis to perform Market Basket Analysis and predictive modelling of item outlet sales. Retail data analysis has evolved from manual sales reporting and basic statistical methods to automated systems capable of uncovering hidden associations in consumer purchasing behavior. Early approaches, such as the Apriori algorithm, laid the groundwork for identifying product cooccurrences. However, as retail data became more complex and voluminous, traditional systems became inadequate. The core problem addressed by this project is the inability of conventional batch-oriented systems to process and analyze large-scale transactional data in real time. This results in delayed insights, excessive manual intervention, and suboptimal business strategies. A dynamic and integrated solution is needed to streamline data processing and leverage machine learning for accurate, actionable predictions. The proposed system combines a robust Django web framework with state-ofthe- art machine learning algorithms-specifically, Random Forest and Decision Tree regressors. It automates the entire process from data upload, cleaning, and preprocessing to model training, prediction generation, and evaluation. Key features include secure user authentication, seamless CSV data uploads, and comprehensive data cleaning, such as missing value imputation, categorical encoding, and outlier handling. A structured workflow splits the data for training and testing, and model persistence enables real-time inference. Advanced visualization tools provide insights into sales performance and customer purchasing patterns. This system transforms raw retail data into valuable intelligence, enhancing inventory management, optimizing marketing strategies, and driving revenue growth. It ensures businesses remain competitive in a digital marketplace.
Journal Article
Robust Shelf Monitoring Using Supervised Learning for Improving On-Shelf Availability in Retail Stores
This paper proposes a method to robustly monitor shelves in retail stores using supervised learning for improving on-shelf availability. To ensure high on-shelf availability, which is a key factor for improving profits in retail stores, we focus on understanding changes in products regarding increases/decreases in product amounts on the shelves. Our method first detects changed regions of products in an image by using background subtraction followed by moving object removal. It then classifies the detected change regions into several classes representing the actual changes on the shelves, such as “product taken (decrease)” and “product replenished/returned (increase)”, by supervised learning using convolutional neural networks. It finally updates the shelf condition representing the presence/absence of products using classification results and computes the product amount visible in the image as on-shelf availability using the updated shelf condition. Three experiments were conducted using two videos captured from a surveillance camera on the ceiling in a real store. Results of the first and second experiments show the effectiveness of the product change classification in our method. Results of the third experiment show that our method achieves a success rate of 89.6% for on-shelf availability when an error margin is within one product. With high accuracy, store clerks can maintain high on-shelf availability, enabling retail stores to increase profits.
Journal Article