Computational & Mathematical Organization Theory
Adaptive Trust and Co-operation: An Agent-Based Simulation Approach
Proceedings of the workshop on Deception, Fraud, and Trust in Agent Societies held during the Autonomous Agents Conference: Trust in Cyber-societies, Integrating the Human and Artificial Perspectives
Parametric interdependence, learning-by-doing, and industrial structure
Computational & Mathematical Organization Theory
| Hi-index | 0.00 |
A model is designed and used to simulate how partners in a supplyrelationship identify and reach a common target in the form of an ideal endproduct. They cooperate fully and share returns. They learn by interaction,as follows. From their different perspectives, they complement each other‘sidentification of the target. They adapt their productive competencies tothe target, in order to conform to demand (quality), and to each other, inorder to achieve efficient complementarity in production (efficiency). Asthey approach the target, their accuracy of identifying the targetincreases. Also, their speed of adaptation increases, and thus they can besaid to be learning by doing. The model allows two different patterns ofacceleration: a routine and a radical type of development. At some distancefrom the target they start to produce. A longer distance from the targetyields earlier returns, but also entails a greater compromise on quality andthereby yields lower returns. Unpredictable changes in market and technologyyield random shifts of the target. In the analysis, the returns from singleand dual sourcing are compared under different parameter settings. Thesimulations show that in line with expectations dual sourcing can be moreadvantageous if development is of the radical type. However, the advantageonly arises if conditions of market and technology are neither too volatilenor too stable.