On reliability issues in artificial
intelligence-directed modelling and simulation
| “In an earlier article Ören (1986a) indicates the relevance of machine learning in simulation .... He also points out that every time there is learning in a knowledge base, the integrity and consistency of the knowledge base need to be assured via relevant syntactic and semantic checks.” | |
| Khoshnevis,
B., Parisay, S. (1993). Machine Learning and Simulation: Application in
Queuing Systems. Simulation, 61: 5 (Nov.), 295-302.
|
|
| “Ören (1987a, 1989) has demonstrated the wide variety of ways in which AI and simulation are error prone, a situation which can be alleviated with a corresponding variety of quality assurance techniques. Ören’s MAGEST advisor (Ören and Tam, 1988) shows how coupling specifications for GEST can be quality assured. Ören and Sheng (1988) show how to achieve ‘built-in’ quality assurance within a modelling and simulation environment using semantic knowledge expressed as rules." (p. 345) | |
| Zeigler,
B.P. (1990). Object-Oriented Simulation with Hierarchical, Modular Models.
Academic Press, Harcourt Brace Jovanovich Publishers, Boston, Mass. 395
p.
|
|
| "This column reviews some of the AI related aspects of the 1988 Summer Computer Simulation Conference in Seattle, July 25-28. - On Tuesday morning Professor Tuncer I. Oren of the University of Ottawa addressed a plenary session on 'Quality Assurance in the AI Era.' - Professor Oren began his talk by asking for a more practical goal that would also be a major step toward proving correctness. He asked how to write programs and to design models that are able to be audited and, in fact, provide automatic audit trails. He first introduced this concept in an article last year in SIMULATION which discussed knowledge base auditing." (p. 207). | |
| Wildberger,
M. (1988). AI and Simulation. Simulation 51:5 (Nov.), 207.
|
|
| "GEST [9] typifies the state-of-the-art of component models in continuous simulation language. MAGEST performs checks and certifies GEST programs. Thus any GEST module in a model base can be certified to be correct. Similarly, TESS [18] provides support for module development with SLAM networks." (pp. 113-114). | |
| Standridge,
C.R. (1986). An Approach to Model Composition from Existing Models. In:
Modelling and Simulation Methodology in the Artificial Intelligence era,
M.S. Elzas et al. (Eds.), North-Holland, Amsterdam, pp. 113-120.
|
|
| "Yet as the complexity of the problems we leave in our wake gets larger, we may find Ören's perspective on model realism more and more attractive in extricating ourselves from what our limited methodologies have wrought." (p. 220). | |
| Zeigler,
B.P. (1983). Respondent's Comments on T.I. Ören. In: Adequate Modelling
of Systems, H. Wedde (Ed.). Springer-Verlag, W. Germany, p. 220. |
|
| "To increase significantly the probability of success in conducting a simulation study, an organization must have a department or group called simulation quality assurance (SQA). The SQA group is responsible for total quality management and works closely with the simulation project managers in planning, preparing, and administering quality assurance activities throughout the simulation study. The SQA is a managerial approach that is critically essential for the success of a simulation study. Ören [1-3] presents concepts, criteria, and paradigms that can be used in establishing an SQA program within an organization." | |
|
Balci, O. (1998). Verification, Validation, and Testing. In: J. Banks, Editor, Handbook of Simulation, Engineering & Management Press (1998), pp. 335–393. |
|