Search

We present the IBM Research Scenario Planning Advisor (SPA), a decision support system that allows users to generate diverse alternate scenarios of the future and enhance their ability to imagine the different possible outcomes, including unlikely but potentially impactful futures. Our system, takes as input the relevant information from news and social media, representing key risk drivers, as well as the domain knowledge and generates scenarios that explain the key risk drivers and describe the alternative futures. To this end, we provide a characterization of the problem, knowledge engineering methodology, and transformation to AI planning. Furthermore, we describe the computation of the scenarios, lessons learned, and the feedback received from the pilot deployment of the SPA system in IBM. [ABSTRACT FROM AUTHOR]

Domain-Independent planning is known to be a very hard search problem, and in the last three decades many search techniques and heuristics have been developed with the aim of efficiently solving such a task. These techniques and heuristics include the usage of landmarks, which are logical expressions consisting of facts that become true or actions that are executed in any solution plan. We propose the usage of landmarks for speeding up the search of the planner LPG, a system implementing a planning approach based on the use of local search in the space of the action graphs of the planning problem. The results of an experimental evaluation of the proposed techniques show that these techniques can improve the performance of LPG, obtaining a planning system that performs similarly to the state-of-the-art planner LAMA. Moreover, we introduce and experimentally evaluate the concept of "quasi-landmarks"; these are facts that are likely to become true in every solution plan, or facts that must become true in a subset of the solution plans. [ABSTRACT FROM AUTHOR]

Automated planning, which deals with the problem of generating sequences of actions, is an emerging research topic due to its potentially wide range of real-world application domains. As well as developing and improving planning engines, the acquisition of domain-specific knowledge is a promising way to improve the planning process. Domain-specific knowledge can be encoded into the modelling language that a range of planning engines can accept. This makes encoding domain-specific knowledge planner-independent, and entails reformulating the domain models and/or problem specifications. While many encouraging practical results have been derived from such reformulation methods (e.g. learning macro-actions), little attention has been paid to the theoretical properties such as completeness (keeping solvability of reformulated problems). In this paper, we focus on a special case - removing primitive actions replaced by macro-actions. We provide a theoretical study and come up with conditions under which it is safe to remove primitive actions, so completeness of reformulation is preserved. We extend this study also for planning operators (actions are instances of operators). [ABSTRACT FROM AUTHOR]

Description Logic possesses strong knowledge representation and reasoning capabilities and offers logical foundation for Semantic Web ontology languages such as OWL and OWL-S. However, the present implementations of OWL and OWL-S are deficient in semantic modeling for dynamic service composition. They also do not consider user preferences. AI planning possesses a better modeling capability of action state transformations in planning and provides an effective method for solving planning problem and task decomposition. However, it is limited in knowledge representation and reasoning capabilities. Based on merits of Description Logic, OWL-S and AI planning, this paper extends OWL-S model, proposes a service composition mechanism and testifies its feasibility in Description Logic. The results show that this composition mechanism is not only feasible but also helpful for semantic modeling of service composition in Semantic Web. [ABSTRACT FROM AUTHOR]

Workflow management in the grid era faces many challenges both related to workflow modeling and workflow execution. To deal with the challenges, the paper argues that the traditional process-driven workflow management paradigm should be replaced by a goal-driven one and proposes a corresponding implementation approach. The contributions of the paper are as follows: (1) a concept of process pattern is put forward that is an effective means not only for procedural knowledge representation and sharing but also for expertise exchange between domain experts and computer scientists; (2) based on process patterns, a goal-driven workflow framework is designed that incorporates semantic Web technologies and artificial intelligent (AI) planning techniques and that can (semi-)automatically generate a workflow on the fly according to the requirements of users and the running context. [ABSTRACT FROM AUTHOR]