Leo Brecht

Evaluating existing information systems from a business process perspective

The evaluation of existing information systems has gained importance in information systems management. A research project involving ten major Swiss companies and the University of St. Gallen has developed a method for assessing existing system from a business process perspective. Systems are evaluated at two levels: the business process supported, and the technical and functional quality of the system. Users and business managers participate actively in the evaluation project. Evaluation criteria are derived using a critical success factor approach and a set of generally applicable factors. Measurement is considered central to the evaluation, and is supported by a set of generic measures and a catalogue of further measures. The evaluation aims at initiating improvement actions. The method has been tested successfully in the participating companies.

Time-discrete nonparametric hazard model using panel data

In this contribution a nonparametric estimator for the hazard function will be presented for time-discrete survival analysis. The estimator is derived from a likelihood function based upon time-discrete counting processes. With martingale techniques asymptotic properties of the estimator of the cumulative hazard function are shown. Since we consider a nonparametric approach no exploratory variables are considered in the empirical example. For analyzing the remigrant behavior of different foreign nations (Italy, Yugoslavia, Greece, Spain and Turkey) the Socio-Economic Panel (SOEP) is used as a data basis. The estimations are carried out with a module of PRODISA, a program package developed for the analysis of time-discrete duration and panel data for the nonparametric and (semi)parametric case.

Methods for Semiparametric Regression Analysis of Multivariate Correlated Event-History Data

For the analysis of univariate sojourn- or survival-times hazard-rate models are generally used. These models are also applicable for the regression analysis of multivariate sojourn-times. In this case it is possible to formulate the dependent process as a regression model, in which the independent process is modelled as a time dependent covariate. If one is interested, however, in the correlation structure of multivariate sojourn-times, it is necessary to develop new statistical models. This contribution presents methods to generate multivariate survivor functions and develops a semiparametric estimation procedure for the analysis of multivariate correlated sojourn-times. Further generalizations to handle other multivariate event-history data are sketched and it is shown how the concept applies to time discrete multivariate event data.

Zum Stand der Methodenentwicklung

Wir verdichten nun die Ergebnisse aus Abschnitt 3.2. nochmals und erhalten so einen -von der einzelnen Methode unabhangigen — Uberblick uber den Stand der Methodenentwicklung, differenziert nach 17 Gesichtspunkten.

Integrated Methods and Tools

The methods and tools described in this chapter display a particularly high level of integration. In most cases the integration is restricted to consistency in the modelling approach, i.e. the tools are strongly oriented towards the method’s meta-model and deliverables. In contrast, integration with the method’s procedural model or with the techniques each arise in only one method-tool combination (Manganelli/Klein and RapidRe Toolset, see Section 5.4, Hansen and Extend, see Section 53).

Requirements and Summary of Results

Masys Inc. regards Business Process Redesign as a suitable means of bringing about a substantial improvement in its business performance. In the run-up to the “Masys direct” BPR project various consultants made presentations, each praising their own “incomparably good” approach. At the same time uncertainty is growing within Masys about which interpretation of BPR should be adopted, and whether the Organisation Department would be in a position to undertake BPR projects without the help of external consultants. Specifically, Masys needs quick, straightforward answers to the following questions: • Methods: The BPR methods seem to use diverse terminology to conceal similar contents. How can Masys compare and evaluate BPR methods? What are their fundamental differences? Masys is already using methods for TQM and the introduction of package software. What kind of links exist between these and the new BPR methods? • Tools: Masys is initially impressed with the wide-ranging functionality of the tools on offer. There are doubts, however, about whether complex simulation functions and interfaces with CASE systems are really needed. Consequently, Masys wants to identify the crucial features of their future BPR methods for which tool support seems appropriate. How can it then derive specific requirements for these features, e.g. the style of graphical representation? And by no means of least importance are the costs and benefits of tool use.

Systematic Business Change using BPR

Masys Inc. is a medium-sized company having its head office in California and a European branch office in Switzerland. It specialises in the design and marketing of high quality electronic equipment for private households. The product range extends from home cinema systems to internet computers. These products are characterised firstly by the innovative style of their housings, which are always designed by acclaimed artists and available only in limited editions. Secondly, the products’ technical “innards” always meet the highest standards. Masys reaches this extremely limited, but highly profitable market segment increasingly via direct sales, rather than through retailers. This allows it to offer supplementary services, such as calculating the optimal dimensions for home cinema systems.

Beschreibung ausgewählter Methoden

Die vorliegende Untersuchung beschrankt sich auf Methoden fur die projekthafte, grundlegende Neugestaltung betrieblicher Prozesse. Zur Neugestaltung eines Prozesses kann auch der Aufbau der Prozessfuhrung gehoren. Nicht berucksichtigt wurden dagegen Methoden, die sich alleine mit der kontinuierlichen Prozessfuhrung befassen.

Geschäftsstrategie — Prozeß — Informationssystem

Die Informationstechnik ermoglicht grundlegende neue geschaftliche Losungen. Um diese Potentiale systematisch auszuschopfen, ist eine radikale Uberprufung der Prozesse eines Unternehmens notwendig. Die Betrachtung der Prozesse schafft die notwendige Verbindung zwischen Strategie- und Informationssystem-Entwicklung. Die bisher entwickelten Methoden zur Neugestaltung betrieblicher Ablaufe unterscheiden sich teilweise fundamental; eine einheitliche Konstruktionslehre fur Prozesse hat sich noch nicht herausgebildet. In vielen Unternehmen herrscht deshalb erhebliche Unsicherheit uber das richtige Vorgehen. Ausgehend von den Anforderungen der Partnerunternehmen hat das Kompetenzzentrum Prozesentwicklung eine Methode zur Reorganisation von Prozessen entwickelt, die in Pilotprojekten in den Unternehmen bereits eingesetzt wird. Die Methode basiert auf einem einheitlichen Metamodell und beschreibt Ergebnisse, Techniken, Vorgehen und Rollen.

A New Approach for the Analysis of Multi-Episode Duration Data

This contribution is concerned with the statistical analysis of multi-episode models. Semiparametric estimation procedures based on transformation models are developed for the continuous time case under the hypothesis of independent sojourn-times so that the results of Cuzick (1988), Tsiatis (1990) and Ritov (1990) are generalized to the multivariate situation of multi-episode duration data. On the other hand we relax the assumption of independent episodes and develope a semiparametric estimation procedure for conditional independent sojourn-times. In this case the frailty-concept plays the important part of generating multivariate survivor functions and the estimation is carried out using a likelihood based upon the ranks of the sojourn-times generalized for censored data.