Behavior-based decision making: a tutorial
| Autor: | Mohammad Samadi Gharajeh |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Decision support system
Control and Optimization business.industry Computer science Mechanical Engineering 020208 electrical & electronic engineering Control (management) Complex system Inference 02 engineering and technology Machine learning computer.software_genre Style (sociolinguistics) Knowledge base Control and Systems Engineering Modeling and Simulation 0202 electrical engineering electronic engineering information engineering Decomposition (computer science) 020201 artificial intelligence & image processing Artificial intelligence Electrical and Electronic Engineering business Set (psychology) computer Civil and Structural Engineering |
| Zdroj: | International Journal of Dynamics and Control. 6:1816-1840 |
| ISSN: | 2195-2698 2195-268X |
| DOI: | 10.1007/s40435-018-0412-4 |
| Popis: | There are ingenious characteristics in humanistic behaviors so that they can be utilized by the most developers to design smart and complex systems. This paper proposes a novel, knowledge and learning based method called behavior-based decision making, BBDM, in control and system engineering. It is an expert decision support system containing the learning ability to work based on humanistic behavioral reasoning. The knowledge base is built by the system based on various behavioral styles (e.g., safe) associated to other systems and humans. BBDM uses the knowledge-based information to make appropriate decisions when any desired behavioral style is requested from the system. It specifies a success rate for any desired style based on the obtained knowledge base with the aid of a behavioral inference system. This procedure can be used to select a proper system or human to accomplish a requested job. All operations of the BBDM method are performed by a proposed behavioral decision system, called BDS, which consists of three main units: decomposition, behavioral inference, and composition. The decomposition unit splits any behavioral style into several optional features (e.g., safety). The behavioral aggregation sub-unit aggregates all behavioral styles obtained by the system to define the total behavior. The behavioral inference unit produces a success set for any desired behavioral style. Finally, the composition unit converts success set to success rate to specify the success probability of the desired style. Simulation results show that the proposed method has a high efficiency compared to some of the existing decision-making methods. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink