A Framework of Human Systems Engineering. Группа авторов. Читать онлайн. Newlib. NEWLIB.NET

Автор: Группа авторов
Издательство: John Wiley & Sons Limited
Серия:
Жанр произведения: Техническая литература
Год издания: 0
isbn: 9781119698760
Скачать книгу
the authors computed a single value that aggregated the belief coefficients across the project. The authors calculated the geometric mean of the belief values as shown in Equation 3.2. The geometric mean is commonly used to find a single value when trying to aggregate criteria with different properties:

      where

       x’s are individual belief coefficients for each criterion in the project classifier

Schematic illustration of Bayesian belief network example element.

      The scalar measures were then translated into a logistics distribution curve, i.e. the belief coefficients. The belief coefficients indicate areas of misalignment in beliefs between stakeholders, where the beliefs are not objective measures of reality but instead the best understanding of the beliefs of other stakeholders’ expectations, if direct collection of beliefs is unavailable. The belief coefficients were categorized based on their position of the logistics distribution curve. This categorization gives practitioners an indication of the material weaknesses within the system development environment and profiles where risk may be present yet unidentified. Note that these categorizations will vary for different types of system development project genres.

      3.11.1 Modeling the Project

      There are two major components of designing the social system and technical system in tandem in order for them to work together so that the interaction of social and technical factors creates the conditions for successful organizational performance. IE was used as a measure of environment stability, and belief coefficients were used as a measure of alignment of sociotechnical attributes.

      IE was used as a means of gauging confidence in the conditions for success in a multi‐stakeholder environment. IE calculations provided a measure of project environment stability. Changes in entropy provided a mechanism to examine the project over time and to understand the patterns that model environments for success or failure.

Schematic illustration of the case study project social network.

      As can be seen, the project is composed of stakeholders that must work together to deliver the project. For expediency, calculations were made for the edges connected to the program team/PIO.

      In addition to looking at the overall enterprise modeled as a graph structure, it is important to look at the evolution of the alignment coefficients and the overall alignment coefficient over time. Specifically, it is desirable to determine whether the analysis can assess whether the project will succeed based on current and past behavior. For example, random behavior between measuring periods will probably indicate poor predictability of future states. Correspondingly, reasonably stable behavior is an indicator of future performance. Information theory provides a ready measure of the randomness in the system with IE. IE is defined as

      (3.3)equation

      where k refers to the number of distinct states the system may be in.

      IE was chosen as a measure that is at a maximum when the probability of all possible outcomes is equivalent; in essence there is no information that would allow a more educated prediction. Experimental testing indicates the values of both edges, and the system at large generally will cluster around several peak values. If variance (𝜎) was chosen as a measure, it would obscure this information as it is not sensitive to multiple modes. In fact, it is possible that (𝑥) might increase, while 𝜎 could decrease. Consequently, using variance alone would indicate less disorder and more confidence in the model than would be warranted.

      This methodology is designed to identify areas of risk to successful project execution. Risk is defined here as the belief that the project will be challenged or fail given the alignment coefficient and the IE. Figure 3.2 illustrates the possible outcomes of the IE.

      For this discussion, stability S is defined as follows:

      (3.4)equation

      (3.5) Скачать книгу