In diverse projects, variations in stakeholders’ industry backgrounds, management skills, and knowledge can lead to different interpretations of implementation methods and discussions.
When automating project management activities, it is crucial to address the understanding gaps between stakeholders, such as project managers and AI system engineers, concerning “how” and “to what extent” automation should be implemented.
To bridge these gaps, AI Level Definition were developed as a standardized framework, inspired by the levels of autonomous driving. This framework helps establish a common understanding, regardless of differences in management skills or knowledge.
AI Level Definition provide clear benchmarks for the level of automation achieved in project management tools, categorizing the extent of automation with precise criteria.
The AI Level Definition are organized by different project management knowledge areas, making them comprehensive and easy to understand.
The levels of automation are defined as follows:
Level | Index |
Level 0 | Information management only, without control. |
Level 1 | Operates with simple control algorithms. |
Level 2 | Chooses parameter combinations based on the situation. |
Level 3 | Increases response patterns as the user of the AI specifies learning methods. |
Level 4 | Tests hypotheses and learns continuously from data to enhance response patterns. |
The criteria are explained using clear classifications and concrete examples, ensuring accessibility and comprehensibility for all users.