JP-2026075097-A - Operational control method, system, and program for switching between multiple decision support modes

Abstract

[Problem] In decision support systems, continuously using a single decision support method despite changes in evaluator circumstances, time constraints, risk tolerance, etc., may improve efficiency in the short term, but in the long term, it leads to rigid judgment and an increase in oversights. Furthermore, even when using multiple methods in combination, stable operation becomes difficult if the switching between them is subjective or arbitrary. [Solution] The present invention provides an operational control method, system, and program that select an appropriate mode from among a plurality of decision support modes based on at least one of evaluator conditions, time constraints, or risk indicators, and executes decision support processing according to the selection result, thereby realizing flexible decision support that balances short-term efficiency and long-term stability. [Selection Diagram] Figure 1

Inventors

• 望月 浩二

Assignees

• 望月 浩二

Dates

Publication Date

20260507

Application Date

20260106

Claims (1)

1. A method for controlling the operation of a decision support system that has multiple decision support modes, Based on evaluator criteria, time constraints, or risk indicators, Select at least one of the aforementioned multiple decision support modes, A method for controlling the operation of a decision support system, characterized by including the step of executing a decision support process based on a selected decision support mode.

Description

This invention relates to the operation control of a decision support system. This relates to a technology that switches between multiple decision support modes based on conditions. In recent years, decision support systems using artificial intelligence have seen increasing use in diverse fields such as corporate management, research and development, design support, and administrative decision-making. These systems aim to provide support information for evaluators to make decisions by generating multiple options and decision-making factors using search algorithms and inference models. Conventionally, in such decision support systems, methods for improving search speed and accuracy are known, such as parallel processing to increase computing resources and ensemble learning which uses multiple models in combination. However, in all of these methods, the search structure and evaluation criteria themselves remain fixed, making it difficult to flexibly respond to changes in the evaluator's situation, judgment criteria, or operational objectives. Furthermore, it has been pointed out that while continuous use of a single decision support method may improve processing efficiency in the short term, it can lead to problems such as the search results becoming overly tailored to the evaluator's preferences and past judgments, resulting in increased oversights and rigid decision-making. On the other hand, even when using multiple decision support methods in combination, if the switching between them is not based on clear criteria and is done subjectively or arbitrarily, there is a problem of reduced reproducibility and explainability of the operation. Thus, there is a need for a systematic and controllable operational technique to appropriately switch between decision support methods depending on operational conditions such as evaluator conditions, time constraints, or risk tolerance, but this has not been adequately achieved with conventional technology. This is a block diagram showing the functional configuration of a decision support system according to one embodiment of the present invention, illustrating the relationship between a condition acquisition unit that acquires operating conditions, a mode selection unit that selects a decision support mode based on said operating conditions, and a decision support processing unit that executes decision support processing according to the selected decision support mode. The embodiments for carrying out the present invention will be described below with reference to the drawings. However, the present invention is not limited to the embodiments shown below, and various modifications can be made without departing from its essence. (Embodiment 1) The decision support system according to this embodiment comprises at least a condition acquisition unit, a mode selection unit, and a decision support processing unit. The condition acquisition unit acquires operational conditions for providing decision support. These operational conditions may include evaluator conditions such as the evaluator's judgment criteria, preferences, role, or past decision history; time constraints indicating the time allowed for decision-making; and at least one risk indicator indicating the degree of impact or tolerance for judgment errors. The mode selection unit selects an appropriate mode from among multiple decision support modes based on the operating conditions acquired by the condition acquisition unit. The decision support modes may include a high-speed mode that presents results suitable for the evaluator in a short time, a sustained operation mode that emphasizes exploration diversity and stability, or a mode that combines these. The decision support processing unit executes at least one of the following processes: search processing, inference processing, and result presentation processing, according to the decision support mode selected by the mode selection unit. (Embodiment 2) In the above embodiment 1, the mode selection unit may select a decision support mode based on an evaluation function that weights and evaluates multiple operating conditions, or based on predetermined decision rules. In this case, the importance of each operating condition is set according to the system administrator or the operating policy. (Embodiment 3) In the above embodiment 1 or 2, the decision support processing unit may change the range of the search space, the depth of the search, the constraints on the search path, or the termination conditions of the process depending on the selected decision support mode. (Embodiment 4) The decision support system may be configured as a single information processing device, or it may be configured in a distributed manner using multiple information processing devices connected via a communication network. (Embodiment 5) The present invention may be provided as a computer-executable program, and may be implemented by recording the program on a recording medium or distributi