Developing a Formal Model for Cooperative Sourcing


Based on the literature research in the preceding parts, this chapter introduces a formal model of cooperative sourcing (referred to as cooperative sourcing model — CSM) which allows for both analytical and simulative studies in the remainder. First, based on the previous summary of related literature on mathematical outsourcing models (section 2.2.3), the motivation behind choosing this formal approach is explained (section 4.1). Based on the theoretical foundation (chapter 0), the model is successively developed — from different cost functions (process costs and transaction costs) to the cooperative sourcing decision calculus. After providing the basic structure in section 4.2, we distinguish a centralized perspective (4.3) and a decentralized perspective (4.4), completing the model by decision calculi either of the central planner (binary non-linear optimization problem) or the autonomously deciding agents (maximizing individual benefits from cooperative sourcing with uncertainty about the partners’ behavior). Finally, section 4.5 extends the model by considering legal and regulatory constraints which are specific to the banking environment.


Transaction Cost Business Process Agency Cost Decision Function Coalition Member 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 122.
    The model has been discussed and endorsed by the Doctoral Consortium of IRMA International Conference 2005 in San Diego (Beimborn 2005; Best Doctoral Submission Award) and on the 39th Hawaii International Conference on System Sciences 2006 (Beimborn 2006).Google Scholar
  2. 125.
    This measure was applied in the context of social network analysis (Alstyne and Brynjolfsson 1997) and in information retrieval (Salton 1971), for example. For a more detailed discussion of similarity measures, see (Jones and Furnas 1987).Google Scholar
  3. 126.
    Fuzzy set theory, which comes from Zadeh, modifies traditional set theory by defining an element’s membership within a set using a membership function instead of a binary value. Thus, an element belongs to a certain set with a particular degree of between 0 and 1 (Zadeh 1965).Google Scholar
  4. 130.
    Some models make a case that increasing standardization results in decreasing specificity and, thus, strategic value (e.g. Lammers 2004). In contrast, our model assumes no interrelation between similarity degree and basic specificity (represented by the residual core competence measure λtk.Google Scholar
  5. 131.
    Some models (e.g. Tsang 2000, Lammers 2005) include the “strategic value” of a decision in the agent’s objective without explicitly stating how to handle the monetarization. Our model tries to prevent this critical issue by solely incorporating cost aspects, as done in many other models (e.g. Aksin et al. 2004, Knolmayer 1993). This limitation is discussed in section 6.4.Google Scholar
  6. 132.
    The reconsideration of uncertainty within the decision calculus presents a major issue, rad might be determined via the CAPM (Brealey and Myers 1996), if investment projects with identical risk structures could be valued by the capital market (Lammers 2004). Nevertheless, the outcomes of a decision and the associated probabilities must be known. This is a quite rigorous assumption and furthermore reduces the multidimensional risk construct to an expected value of cost outcomes “and the problem becomes a straightforward cost trade-off” (Jurison 1995, 243). For a decision model on IT outsourcing explicitly taking risk into account, nevertheless also in a simplified qualitative way, see e.g. (Jurison 1995).Google Scholar
  7. 135.
    By comparing a centrally and a decentrally coordinated standardization model, Weitzel explores the circumstances in which independent agents’ local decisions about adopting a communication standard lead to an inferior network configuration compared to the optimal solution established by a central planner (Weitzel 2004; Weitzel et al. 2006).Google Scholar

Copyright information

© Betriebswirtschaftlicher Verlag Dr.Th. Gabler | GWV Fachverlage GmbH, Wiesbaden 2008

Personalised recommendations