Research

Research Fields

Members of the department primarily conduct research in the areas of E-Business Standardization, B2B Data Exchange, Management of Product Catalog Data, and E-Goverment. The department's strength has been the close and long-term relationships with companies and institutions from industy as well as academia.

Product Catalog Data

Catalog data possesses a substantial meaning for suppliers. It describes their products and is an instrument for differentiation between the competitors. The task of catalog creation often makes the introduction of new or the extension of existing information systems necessary. A reason is that catalog data is stored in different and distributed operational information systems. Likewise the relevant data is administrated by different organizational units. In addition catalog data management has to integrate several sources and must be able to create catalogs in any XML standard fast and economically. The target systems must be able to import any XML catalog document. Especially for market places, which process hundreds of supplier catalogs, the catalog import is a key task; particularly since it cannot be assumed that all catalogs use the same format and that their quality is evenly high. Therefore catalog data passes through a staging process that covers different technical and semantic checks, operations and release steps. Defined staging processes are useful, because a syntactical correctly created catalog does not guarantee that the catalog corresponds also to content requirements. Concerning this, a statement can be made only in dependence on the respective, often customized requirements.

E-Business Standards for Electronic Product Catalogs

For the exchange of catalog data a number of XML-based standards are available. Catalog standards have to be seen in the context of B2B standardization. On the basis of a level model, standardizations can be partitioned. Often the levels framework, processes, documents, vocabulary and datatypes are formed. Catalog standards define catalog documents, which consist of a vocabulary. The vocabulary contains the elementary data objects, which are specified up to the datatype level. The highest level "processes" is only partially covered by catalog standards. A process is an admissible sequence of documents, e.g. order, order confirmation, delivery notice, invoice. Therefore a catalog process could cover: catalog request, catalog, and catalog update. Finally the level "framework" contains definitions regarding transmission and communication protocols. With reference to the level model the following groups of applicable standards can be formed: The group of genuine catalog standards contains those standards, whose origin is situated in the specification of catalog documents for e-procurement. To this group belong e.g. BMEcat, cXML and eCX. Meanwhile cXML expanded its scope to further business messages; BMEcat is supplemented by the transaction standard openTRANS. Transaction standards go a step further
in standardizing a multiplicity of business messages; catalog documents are just a part of it. Prominent members of this group are EAN.UCC, OAGIS and xCBL. The third group consists of e-business frameworks, which standardize a complete data and communication infrastructure. The most well-known frameworks are ebXML and RosettaNet.

Complex Product & Price Models

Buying companies make high demand on the semantics and syntax of product data, in order to support buyers and other employees in their business functions and needs. Likewise it is a vital interest of suppliers, thus the companies that create electronic catalogs, to be able to describe their products in full detail, according to requirements of customers and in a manner that supports and influences procurement decisions. The product description plays a major role in sales and procurement systems. The main object of catalog-based transaction systems are standardized products of a limited specification and complexity. Among these products are primarily indirect goods that are not an immediate input factor for production processes and can not be attributed to manufactured final goods. A common term is MRO goods (maintenance, repair and operations). Indirect goods are characterized by a limited specification, low single values and high order frequencies as well as at the same time a low share in the procurement budget. Although they require a relevant amount of resources for procurement, order and stock receipt management. The described restric-tions cause a limited area of application for e-procurement systems so far. By extending the capabilities of catalog applications concerning product complexity, product models and product data exchange, e-procurement system could re-shape their role as tools for buying direct, complex or strategic goods as well.

Product Classification Systems

Standardized product classification systems play a major role for searching and comparing offered products on electronic markets. Especially in case of large multi-vendor product catalogs classified data becomes an important asset and success factor. The most known systems are UNSPSC and eCl@ss, however they are still developing, and new systems are emerging as well. Classification systems differ not only in content but also in structure from each other. The task of product classification is to assign each product to a product group corresponding to common attributes or application areas. Though classification systems are not a new phenomenon of B2B e-commerce; they are already in use as an instrument of structuring since decades. The field of application is very broad. It ranges from manufacturing, costing and sale (e.g. product catalogs) up to national and international economic statistics. In B2B e-commerce classification systems gain a new meaning and function. They are an instrument for accessing large e-catalogs. Standardized and supplier-independent classification is an elementary requirement for efficient product search and qualified comparison of products in electronic markets and other catalog-based procurement systems. To describe products in a uniform manner, some classification systems define so-called sets of attributes. A set of attributes is assigned to a classification group and contains the necessary product attributes. In e-catalogs that claim to support the classification system each product has be described by the group-depend set of attributes

XML Schema Languages in B2B

Since the advent of XML as an universal language for describing data on the web, data exchange in business-to-business relationships has to answer the question how to close the gap between relational and hierarchical representation of the same of similar information. One important aspect is the way in which we describe the syntax - and if possible - the semantics of the data. The transformation of data from a relational database into a standardized XML document and back into another relational database is essentially influenced by the formal specification of the data structures and its quality. While a relational database is described precisely by its conceptual schema, such a specification for XML documents depends mainly on the capabilities of the selected schema language. So far genuine XML database systems are hardly used for e-business applications; therefore the transformation of XML documents into relational databases (and in reverse) is still a main task in electronic data interchange between enterprises. A substantial reason is that e-business systems connect existing operational information systems, which are based almost exclusively on relational models and database systems. In order to keep XML documents in relational databases persistent, it is necessary to define a database schema that permits the representation of content and structures of XML files as loss-free as possible. Inlining methods point out that such a transformation of documents, which are specified in a XML schema language, is possible and thus storage in relational databases can be realized. However, the quality of the transformation, especially regarding the implicit semantics, depends on the meta information that is formalized in the specification of the catalog standard. Newer XML schema languages can express more semantic information, e.g. datatypes and relational concepts, which facilitate the transformation process or even enable their loss-free execution.