Abstract
Although on-line shops have become very popular lately, there are still some sectors of industry, such as clothing and furniture, where the on-line shop concept has not penetrated the market yet. The main reason for that is the diversity of the specific sectors, the mentality of the market players and the lack of user friendly environments, where the customer could actually “see” the products. This paper presents an Intelligent Configurable Electronic Shop Platform, based on Ontologies and 3D Visualization, that aims at enabling a suitable representation of products with the most realistic possible visualization outcome. The platform, designed for the furniture sector, includes all the practicable electronic commerce variants and its on-line product configuration process is controlled by an ontology that was created using the OWL Web Ontology Language.
Introduction
While e-commerce has flourished for standardized goods like books or electronics, sectors like furniture and clothing have lagged. This delay is primarily due to high product diversity, where items are highly customizable (different fabrics, legs, dimensions), and a significant "visualization gap" that makes customers hesitant to purchase expensive items without a realistic sense of the final assembly. Furthermore, many existing e-shops suffer from being disconnected from internal Enterprise Resource Planning (ERP) systems, leading to data silos.
The INCOVIS (Intelligent Configurable Electronic Shop) platform addresses these challenges by decoupling product "logic"—governed by an intelligent ontology—from the "look," handled by an interactive 3D environment.
System Architecture and Methodology
The INCOVIS platform is built on a modular architecture designed to handle complex product constraints and real-time visualization.
At the core of the system is an Ontology-Driven Configurator. Instead of hard-coding product rules, the system utilizes OWL-DL (Web Ontology Language - Description Logic). This allows the system to reason about product constraints dynamically. For instance, if a user selects a specific chair frame, the ontology automatically restricts available seat and leg options based on compatibility and price rules defined by the manufacturer.
The front end leverages VRML (Virtual Reality Modeling Language) and the External Authoring Interface (EAI) to enable communication between the web browser and the 3D scene. A unique Animation Unit records and replays assembly steps, helping users understand the construction of their configured product. The backend is powered by Java Servlets on Apache Tomcat, using the Jena framework for semantic reasoning and MySQL for data persistence.
Implementation and Data
The platform was evaluated using a proprietary dataset of furniture products provided by DIFROS S.A., a furniture manufacturing company. This dataset included 3D models of components (legs, seats, backs) along with metadata regarding materials, pricing, and compatibility constraints. To ensure business viability, the authors implemented an "Off-Line Staging" mechanism to synchronize e-shop data with the company's legacy ERP systems, balancing security and infrastructure costs.
Key Insights: Logic vs. Presentation
A central innovation of this work is the separation of product logic from the user interface. By encoding the "rules" of furniture assembly into an OWL ontology using cardinality and value restrictions, developers can update product lines without modifying the underlying software code. This constraint-based approach prevents users from configuring non-manufacturable products, such as a chair with missing components or incompatible materials.
Key Results
The functional prototype demonstrated a seamless web-based interface where users could select parts and observe them "snap" together in a 3D environment while tracking total costs in real-time.
The pilot with DIFROS S.A. validated that the platform could handle real-world inventory and present it realistically enough for commercial application. Additionally, the synchronization algorithm proved efficient for maintaining data consistency between the e-shop and legacy ERP systems.
Conclusion and Future Work
While the platform relies on now-superseded technologies like VRML, the underlying principles of ontology-driven configuration remain highly relevant. Future developments could involve migrating to modern frameworks like WebGL/X3D and automating ontology creation directly from manufacturing databases. Despite its qualitative evaluation, the INCOVIS platform provides a robust blueprint for bridging the gap in complex online retail sectors.
Summary updated