Many System Dynamics researchers have found that decision-makers have difficulties in controlling System Dynamics model which represent complex social reality. This means that heuristics employed by decision-markers are not appropriate for controlling dynamic social problems. As alternative ways for understanding and controlling System Dynamics models, various mathematical methods have been suggested. Some simulation-based experiment demonstrated the possibility of decision-makers’ learning ability. For instance, the experiment performed by Sterman showed that game players' performance was improved slowly as their experiences are accumulated. The slow learning process is often regarded as indicating the limitation of human intelligence. On the contrary, it may be interpreted as indicating a potential power of human intelligence or heuristics. In previous studies, decision-makers’ heuristics are formulated in simple decision rule. Such decisions rules failed to incorporate the learning ability of decision-makers. To experiment the learning ability of decision-makers, this study replace decision-makers with neural network model. The neural networks are recognized as a representative of human intelligence by many students in artificial intelligence. In this study, neural network heuristic are applied to two System Dynamics models; Meadow's commodity cycle (1969) and Sterman’s model of the Kondratiev cycle, or long wave (1985). Neural networks model have demonstrated a surprising performance in learning and pattern recognition. In addition to neural networks applications, this study demonstrated technical feasibilities in IBM environments using Smalltalk.

This is the whole item.

Date created

• 1994

Type

• Document

Processing Activity

• Processing documentation available at: https://wiki.albany.edu/display/SCA/Processing+Ingested+Digital+Files

License

• Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International

  
  Contact Us for questions or additional rights.

ITEM CONTEXT

Part of

fe35db792b573af835d96e6eba4759cd

Scope and Contents

Part of

c060552994c1527f70693734935660f1

Scope and Contents

Part of

23d738ba88f8333bc39725f9cb5bd0b8

Scope and Contents

Collection

System Dynamic Society Records

Scope and Contents

Collecting area

• This collection is part of the University Archives