University of the Aegean
Estimation of consumers’ preferences and demand for integrated e-shopping and parcel delivery services applying discrete choice models and their incorporation in management business tools for the market sector of postal services
Abstract
dc:descriptionThis research explores how digital technologies and data reshape e-shopping and parcel delivery services. It aims to analyze consumer preferences to design a Data-Driven Business Model (DDBM) for last-mile logistics services. The study uses Discrete Choice Analysis to understand factors influencing consumer choices, such as (1) delivery place, (2) time and speed, (3) track and trace, (4) value-added services, and (5) delivery cost. The literature review's research drivers provided insights into designing our methodology to measure consumers’ preferences. We have applied the framework of Design Science Research (DSR), which is used as an essential and legitimate research framework in Information Systems (IS) research and helps managers and IS specialists to express the business logic of a firm. To collect data we developed a structured questionnaire consisting of two main sections. The first one refers to the Revealed Reference questions and the second one to the Stated Preference Experiments (SPs). We estimated four different types of Discrete Choice Models. First, an MNL model was estimated to measure consumers’ preferences between five service alternatives (including the No-Choice option) and operate as a reference model. Then we applied an N-MNL model due to the ‘No-choice’ option which violates the property of Independence of Irrelevant Alternatives (IIA). The presence of multiple observations on stated choice responses for each sampled individual in the Questionnaire showed correlated responses across observations which is a violation of the independence of observations assumption in the standard choice model, thus we estimated a mixed logit model (M-LM). The willingness to pay (WTP) was calculated based on the M-LM and it was found that customers are willing to pay 0.26€ for the products to be delivered to their home, and 0.78€ for delivery at a specific time window but not on ‘Saturday morning’ as this service level did not attain any significance. To incorporate the behavioral elements into the discrete choice model, we proceeded with the estimation of a Hybrid Choice Model (HCM). The findings from the four discrete choice models indicate that consumers prefer faster deliveries and are willing to pay extra for specific delivery options, but not for additional services. We have also estimated three clusters, based on the results of the HCB model, applying the k-means algorithm, presenting the predicted probabilities for each service alternative to be chosen by each customer and choice task.The research proposes a DDBM for Delivery Service Providers (DSPs) focusing on digital platforms and efficient supply chains. It uses the VISOR framework to develop a taxonomy for last-mile logistics services, applied to the Hellenic Post in Greece. The model includes economic, environmental, and social KPIs, aligning with the Triple Bottom Line concept. Overall, the study contributes to understanding consumer behavior in e-shopping and delivery services, providing insights for businesses to adapt and improve customer experience in a rapidly changing market.
Degree
thesis:*- Grantor dc:publisher
- University of the Aegean
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Irakleous, Alexia
- Ηρακλέους, Αλεξία
Subjects
dc:subject × 26- Ηλεκτρονικές αγορές
- Yπηρεσία παράδοσης ηλεκτρονικής παραγγελίας
- Tελευταίο μίλι της εφοδιαστικής αλυσίδας
- Συμπεριφορά καταναλωτή
- Ανάλυση διακριτών επιλογών (ΑΔΕ)
- Πειράματα δεδηλωμένης προτίμησης
- Λανθάνουσες μεταβλητές
- Επιχειρηματικός καμβάς
- Tαχυδρομικές υπηρεσίες
- Επιχειρηματικό μοντέλο καθοδηγούμενο από δεδομένα
- e-shopping
- Parcel delivery services
- Last-mile logistics services
- Consumer choice behavior
- Discrete choice models
- Latent variables
- Stated preference experiments
- Data-driven business model
- VISOR framework
- Postal services
- Οικονομικά και Επιχειρήσεις
- Κοινωνικές Επιστήμες
- Επιστήμη λήψης αποφάσεων
- Economics and Business
- Social Sciences
- Decision Sciences
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/57642