Analysis of Intelligent Decision Support Systems and a Multi Criteria Framework for Assessment

Aijie Wang, Bin LiangJournal of Enterprise and Business Intelligence 2023 · DOI

Abstract

The act of decision-making lies at the core of human existence and shapes our interactions with the surrounding environment. This article investigates the utilization of artificial intelligence (AI) techniques in the advancement of intelligent decision support systems (IDSS). It builds upon prior research conducted in the decision-making field and the subsequent development of decision support systems (DSS) based on that knowledge. The initial establishment of the fundamental principles of classical DSS is undertaken. The subsequent emphasis is directed towards the integration of artificial intelligence techniques within IDSS. The evaluation of an IDSS, as well as any other DSS, is a crucial undertaking in order to gain insights into the system's capabilities and identify areas that require enhancement. This article presents a review conducted on this significant yet insufficiently investigated subject matter. When utilized in conjunction with DSS, AI techniques such as intelligent agents, artificial neural networks (ANN), evolutionary computing, case-based reasoning, and fuzzy logic provide valuable assistance in defining complex practical challenges, which are mostly time-critical, encompass extensive and scattered data, and can derive advantages from sophisticated reasoning.

Introduction: The Evolution of Decision Support Systems

Decision-making is a fundamental human activity, but as data becomes increasingly "scattered" and time-critical, human cognitive limits and biases can become significant bottlenecks. Traditional Decision Support Systems (DSS) were developed to assist with structured problems, such as optimization or routing. However, modern organizational challenges often involve semi-structured problems—like entering a new international market—where data is ambiguous and standard logic-based rules may fail.

The traditional flow of Management Information Systems (MIS) and Decision Support Systems, highlighting the path from data processing to management decisions.

The evolution of these systems has moved through four generations, from early data-centric tools to modern web-based, AI-driven platforms. The goal of an Intelligent Decision Support System (IDSS) is to augment human expertise, particularly in emergency response or financial trading, where rapid processing of vast datasets is required.

AI Integration in IDSS

To handle the ambiguity of semi-structured decisions, IDSS integrate various Artificial Intelligence (AI) techniques that mimic human reasoning. Key methodologies reviewed include:

  • Artificial Neural Networks (ANNs): These are used for pattern recognition and handling non-linear data where traditional logic fails. While powerful, their "black box" nature can be a hurdle for transparency.
  • Fuzzy Logic: This handles "imprecise" human concepts (e.g., "likely" or "very good") rather than binary 0s and 1s, allowing for more natural human-computer interaction.
  • Evolutionary Computing (Genetic Algorithms): This technique uses principles of natural selection, such as crossover and mutation, to find optimal solutions in massive search spaces.
  • Intelligent Agents: These are autonomous software entities that can collaborate or negotiate to reach a specific objective.
A typical evolutionary computing flowchart, illustrating the iterative process of population initialization, fitness assessment, and genetic operators like crossover and mutation.

A Multi-Criteria Assessment Framework

A significant challenge in IDSS development is evaluation. Because these systems are often complex, standard metrics often fail to capture their true utility. The authors propose a structured evaluation framework using the Analytic Hierarchy Process (AHP).

AHP allows managers to organize and analyze complex decisions by breaking them down into a hierarchy. The proposed model maps IDSS requirements into a five-level hierarchy, allowing stakeholders to assign weights to different criteria—such as whether the user interface is more important than processing capacity for a specific task—to calculate a final "Decision Value" score. This framework distinguishes between Process Values (efficiency, proficiency) and Outcome Values (organizational performance).

Key Insights and Conclusion

A central theme of the research is the concept of "Satisficing" over "Maximizing." Drawing on Herbert Simon’s theories, the paper notes that humans and IDSS often aim for solutions that are "good enough" given cognitive and temporal limits, rather than perfect solutions that are computationally impossible to find in real-time.

Furthermore, the paper highlights that an IDSS should be judged not just on the final result, but on how much it reduces the "cognitive load" on the human user. While a "universal" IDSS remains a future goal, the hybridization of techniques—such as combining ANNs with Fuzzy Logic to improve transparency—represents a promising path forward. Future research will likely focus on these hybrid systems to overcome the individual limitations of current AI methodologies.

Summary updated

Abstract

The act of decision-making lies at the core of human existence and shapes our interactions with the surrounding environment. This article investigates the utilization of artificial intelligence (AI) techniques in the advancement of intelligent decision support systems (IDSS). It builds upon prior research conducted in the decision-making field and the subsequent development of decision support systems (DSS) based on that knowledge. The initial establishment of the fundamental principles of classical DSS is undertaken. The subsequent emphasis is directed towards the integration of artificial intelligence techniques within IDSS. The evaluation of an IDSS, as well as any other DSS, is a crucial undertaking in order to gain insights into the system's capabilities and identify areas that require enhancement. This article presents a review conducted on this significant yet insufficiently investigated subject matter. When utilized in conjunction with DSS, AI techniques such as intelligent agents, artificial neural networks (ANN), evolutionary computing, case-based reasoning, and fuzzy logic provide valuable assistance in defining complex practical challenges, which are mostly time-critical, encompass extensive and scattered data, and can derive advantages from sophisticated reasoning.
BibTeX
@article{wang2023analysis,
  author = {Aijie Wang and Bin Liang},
  title = {{Analysis of Intelligent Decision Support Systems and a Multi Criteria Framework for Assessment}},
  journal = {Journal of Enterprise and Business Intelligence},
  year = {2023},
  doi = {10.53759/5181/jebi202303022},
}
Figure