580 California St., Suite 400
San Francisco, CA, 94104
Key research themes
1. How does automation and human collaboration influence dynamic decision-making responsibility and outcomes?
This area investigates the interaction between human decision makers and automated or AI-driven systems in dynamic, non-static decision-making contexts. These studies focus on quantifying how automation affects human influence and causal responsibility over time, often with implications for the design of supervisory systems and trust in human-automation teams. Understanding the dynamics of influence aids in assessing when automation should support versus override human decisions, with considerations for sensitivity, detection capability, and human adaptability.
Key finding: Introduces a novel model quantifying human and automation influence in dynamic decision events using causation strength; finds that automation must demonstrate high detection sensitivity to meaningfully influence human... Read more
Key finding: Differentiates human decision-making into diagnosis and look-ahead reasoning, emphasizing that decision support systems (DMSS) predominantly perform look-ahead functions via heuristic search and 'what-if' analysis. This... Read more
Key finding: Evaluates how computer-aided decision support compensates for human cognitive limitations such as information processing capacity, highlighting that humans often use satisficing strategies due to these constraints. The paper... Read more
2. What are the frameworks and architectural considerations for intelligent decision-making support systems (i-DMSS) in organizational contexts?
This research stream explores foundational and integrative design frameworks underpinning intelligent computerized systems that aid or automate decision-making in organizations. It includes the evolution of decision support paradigms, system architectures combining data, knowledge, and human factors, and examines the computational mechanisms needed to effectively support complex, adaptive organizational decisions. These frameworks are crucial to bridging cognitive models and computational implementation for practical and scalable i-DMSS.
Key finding: Reviews five major design frameworks for intelligent decision-making support systems, highlighting the lack of an integrated standard architecture that links decision-making processes with computational mechanisms. It... Read more
Key finding: Reassesses and revitalizes the concept of organizational decision systems, tracing their evolution since the 1960s, differentiating between decision automation and decision support systems. Predicts embedding of analytics and... Read more
3. How can decision-theoretic goal models and machine learning enhance decision-making under uncertainty and dynamic environments?
Focused on modeling and reasoning with probabilistic and utility-aware goal representations, this theme explores approaches to capture uncertainty, preferences, and dynamic context evolution in decision processes. Decision-theoretic goal models extend traditional requirement engineering by incorporating probabilities and utilities, enabling automated reasoning to identify optimal action plans. Complementing this, machine learning-driven rule evolution supports decision adaptation in dynamic environments, enhancing responsiveness and reducing human effort.
Key finding: Introduces decision-theoretic goals integrating probabilistic effects of actions and stakeholder utility functions into goal models, enabling the calculation of optimal action plans with a state-of-the-art reasoning tool... Read more
Key finding: Proposes a component for continuous evolution of rule knowledge bases at runtime by leveraging hybrid machine learning (decision trees) and genetic algorithms to generate and optimize decision rules in dynamic contexts.... Read more