Time filter

Source Type

Nürnberg, Germany

Roser S.,Senacor Technologies AG | Muller J.P.,Clausthal University of Technology | Bauer B.,University of Augsburg
Information Systems and e-Business Management | Year: 2011

Our work aims at providing support for the decision-making processes involved in the model-driven development of information technology (IT) solutions for cross-organizational business process (CBP) coordination and automation. The objective of the work described in this paper is to provide enterprise IT architects with an evaluation and decision model that enables the principled assessment and selection of an effective IT architecture paradigm (e. g. central broker, federated brokers, peer-to-peer) for a given cross-organizational business process coordination task. Our approach follows the principles of design science; the contribution of this paper is threefold: First, we present three common architectural patterns for (service-oriented) CBP coordination. Second, the core contribution is established by a new method for decision support suitable for IT architects to derive and evaluate an appropriate architecture paradigm for a given use case or application domain. The method is accompanied by a set of representative scenario descriptions that allow the evaluation and selection of appropriate IT system coordination architecture paradigms for CBP enactment, as well as a set of guidelines for how different contingencies influence IT system coordination architecture. © 2010 Springer-Verlag.

Sebastian S.,University of Mannheim | Stefan B.,Senacor Technologies AG
Proceedings - Pacific Asia Conference on Information Systems, PACIS 2012 | Year: 2012

In the recent years, Software-as-a-Service has gained growing attention from software vendors as well as software customers. In this distribution and pricing model, software vendors are responsible for the operation and maintenance of solutions and customers pay for the service in form of continuous usage-based subscription fees. This study analyses the key characteristics of the Softwareas- a-Service concept and evaluates their influence on development processes of software vendors. Based on the literature, two defining characteristics (vendor-hosted, pay-per-usage) and five supportive characteristics (standardization, web-technologies, multi-tenancy, fine-granularity, continuous evolution) are identified. Two vendors of complex business applications - both applying a mix of deployment models - are analyzed to identify the impact of Software-as-a-Service on development processes, as well as the driving forces behind the change. The results indicate, that the concept especially affects the requirements engineering as well as operations phases. The combination of the defining characteristics results in an increased cost as well as innovation pressure. Vendors are challenged to optimize and streamline existing practices to reduce internal costs. At the same time, vendors need to innovate quicker. The combination of these aspects asks for an increased integration of internal activities like development and operations, as well as an increased customer-orientation and integration.

Kaiser M.,Senacor Technologies AG | Ullrich C.,Research Center
ECIS 2014 Proceedings - 22nd European Conference on Information Systems | Year: 2014

Information systems (IS) projects are famous for experiencing severe cost overruns, which amongst others are often caused by inaccurate ex-ante cost estimations. Against this background, this article presents a descriptive case study located in an IS transformation program at a major German financial services provider. In this case study, a multi-stage cost estimation process, which was applied to 79 IS projects, is described and the estimation accuracy of the cost estimations of all IS projects is determined using different estimation accuracy measures: Estimating Quality Factor, Forecast Error, and Mean Absolute Percentage Error. Depending on the concrete estimation accuracy measure used for the evaluation, the overall estimation quality of the program turns out to be evaluated as good or at least average - which seems to be contrary to most studies in scientific literature. However, the results further reveal that the estimation accuracy also depends on the estimation accuracy measure chosen for the evaluation. These differing judgements are discussed from a management perspective.

Gorz Q.,University of Augsburg | Kaiser M.,Senacor Technologies AG
Lecture Notes in Business Information Processing | Year: 2012

Owing to the fact that insufficient data quality usually leads to wrong decisions and high costs, managing data quality is a prerequisite for the successful execution of business and decision processes. An economics-driven management of data quality is in need of efficient measurement procedures, which allow for a predominantly automated identification of poor data quality. Against this background the paper investigates how metrics for the DQ dimensions completeness, validity, and currency can be aggregated to derive an indicator for accuracy. Therefore existing approaches to measure these dimensions are analyzed in order to make explicit, which metric addresses which aspect of data quality. Based on this analysis, an indicator function is designed returning a measure for accuracy on different levels of a data resource. The indicator function's applicability is demonstrated using a customer database example. © 2012 Springer-Verlag Berlin Heidelberg.

Kramer T.,University of Mannheim | Eschweiler M.,Senacor Technologies AG
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2013

This paper seeks to address the decision making problem in software development outsourcing scenarios in which a project manager is in charge of deciding about which software components will be outsourced and which ones will be developed internally. Therefore we propose a methodology and tool support which leverage the classification of a project's software components by means of a graph-based model of the components' requirements and their corresponding clustering. In the course of our design oriented research approach, a prototypical implementation of the methodology has been developed and evaluated. It illustrates the practical applicability of the proposed method. We thereby contribute to the location selection problem in distributed software projects and give guidance for in-house or external software production. The theoretical contribution consists of revealing an improved processing methodology for assessing software requirements and increasing the outsourcing success of a software project. Our contribution for practice is an implemented prototype for project leads of distributed teams. © 2013 Springer-Verlag.

Discover hidden collaborations