Learnable Representation for Real World Planning
نویسندگان
چکیده
This paper presents a learnable representation for real-world planning systems. This representation is a significant extension of the ones used in the most recent systems from the Disciple family, the Disciple-Workaround system for plan generation, and the Disciple-COA system for plan critiquing. This representation is defined to support an integration of domain modeling, knowledge acquisition, learning and planning, in a mixed-initiative framework. It also helps to remove the current distinction between the development phase of a planning system and its maintenance phase. It provides an elegant solution to the knowledge expressiveness / knowledge efficiency trade-off, and allows reasoning with incomplete or partially incorrect knowledge. These qualities of the representation are supported by several experimental results.
منابع مشابه
Integrated Process Planning and Active Scheduling in a Supply Chain-A Learnable Architecture Approach
Through the lens of supply chain management, integrating process planning decisions and scheduling plans becomes an issue of great challenge and importance. Dealing with the problem paves the way to devising operation schedules with minimum makespan; considering the flexible process sequences, it can be viewed as a fundamental tool for achieving the scheme, too. To deal with this integration, t...
متن کاملSolving a generalized aggregate production planning problem by genetic algorithms
This paper presents a genetic algorithm (GA) for solving a generalized model of single-item resource-constrained aggregate production planning (APP) with linear cost functions. APP belongs to a class of pro-duction planning problems in which there is a single production variable representing the total production of all products. We linearize a linear mixed-integer model of APP subject to hiring...
متن کاملAutonomous Learning of Action Models for Planning
This paper introduces two new frameworks for learning action models for planning. In the mistake-bounded planning framework, the learner has access to a planner for the given model representation, a simulator, and a planning problem generator, and aims to learn a model with at most a polynomial number of faulty plans. In the planned exploration framework, the learner does not have access to a p...
متن کاملLearnable Evolutionary Optimization in Autonomous Pickup & Delivery Planning
This report describes an evolutionary approach to distributed planning by transport agents. These agents, representing trucks, autonomously decide on their transport orders by interacting with cargo agents representing containers. The agents use a guided evolutionary computation method, called the learnable evolution model, to create transport plans and render optimized decisions on which cargo...
متن کاملA MULTISTAGE EXPANSION PLANNING METHOD FOR OPTIMAL SUBSTATION PLACEMENT
Optimal distribution substation placement is one of the major components of optimal distribution system planning projects. In this paper optimal substation placement problem is solved using Imperialist Competitive Algorithm (ICA) as a new developed heuristic optimization algorithm. This procedure gives the optimal size, site and installation time of medium voltage substation, using their relate...
متن کامل