This paper postulates a rigorous method for the construction of classical planning domain models. We describe, with the help of a non-trivial example, a tool supported method for encoding such models. The method results in an `object-centred' specification of the domain that lifts the representation from the level of the literal to the level of the object. Thus, for example, operators are defined in terms of how they change the state of objects, and planning states are defined as amalgams of the objects' states. The method features two classes of tools: for initial capture and validation of the domain model; and for operationalising the domain model (a process we call compilation) for later planning. Here we focus on compilation tools used to generate macros and goal orders to be utilised at plan generation time. We describe them in depth, and evaluate empirically their combined benefits in plan-generation speed-up. The method's main benefit is in helping the modeller to pro...

This paper describes our experience in capturing, using a formal specification language, a model of the knowledge-intensive domain of oceanic air traffic control. This model is intended to form part of the requirements specification for a decision support system for air traffic controllers. We give an overview of the methods we used in analysing the scope of the domain, choosing an appropriate formalism, developing a domain model, and validating the model in various ways. Central to the method was the development of a formal requirements engineering environment which provided automated tools for model validation and maintenance

Because of the growing complexity, the design of and reasoning about modern industrial systems becomes increasingly difficult. In order to understand and estimate the dynamic system behaviour, appropriate models have to be used. In many cases, existing mathematical models like queuing networks, Markov chain models, or perturbation analysis cannot be applied. In such cases, usually a model is constructed that can be validated by means of computer simulation. Since industrial systems exhibit concurrency, formalisms developed to reason about concurrent systems are also well suited for developing models in this specific application area. Models of systems can be expressed, for instance, in terms of Petri nets or in terms of programs written in a concurrent programming language, like Timed CSP. Both approaches, originating from computer science, are increasingly often applied in modelling of manufacturing systems. In this paper, we present a simple modular approach to the specification of (...

ion In Planning Domain Theories T. L. McCluskey lee@zeus.hud.ac.uk J. M. Porteous julie@zeus.hud.ac.uk The School of Computing and Mathematics, The University of Huddersfield, Queensgate, Huddersfield HD1 3DH, UK Abstract This paper introduces a new planning algorithm which relies on a set of iterative macros and a set of necessary goal ordering constraints that are generated during a compilation phase prior to planning. The main contribution of domain compilation is to speed up planning and is achieved at the price of a well specified domain model that has been engineered into a sort abstracted form. This means that object classes have been identified and object states specified by means of state invariants. A typical planning state corresponds therefore to a conjunction of separate object states. Rather than being a burden for real applications, the use of a structured domain specification, and domain processing tools, is consistent with proper validation of the planning model ...

. This document presents a steam-boiler control specification for the seminar "Semantische Methoden im Wettbewerb" on June 5-9 1995 in Schloss Dagstuhl, Germany based on the informal specification provided by Jean-Raymond Abrial ([2]). The style of the specification itself is strongly related to VDM. Therefore most of the definitions are explicit as they used to be in traditional VDM specifications. The specification language used is RSL (the RAISE Specification Language [1]). 1 Introduction RAISE (which stands for "Rigorous Approach to Industrial Software Engineering ") has been developed under the ESPRIT programme founded by the European Community. Its main aim was to provide a means of creating more reliable and correct software. RAISE consists of two parts: 1. A development method based on the stepwise refinement paradigm ("invent and verify"). 2. A wide spectrum specification language, the RAISE Specification Language (RSL). Throughout the development process only RSL notation is...

This paper presents an approach to formal specification of task management models for interactive systems. The approach is well suited to data-intensive applications in which the system is being used to manage complex collections of interrelated objects. The approach consists of annotating objects with status information, and relating status back to properties of the underlying collection. Status information is used to guide and control the application of activities. The paper illustrates the approach on an example from interactive theorem proving.

Although originally developed for modelling and analysis of computer hardware and software, Petri Nets and CSP are also well suited to model information and control flows in manufacturing systems usually exhibiting concurrency and nondeterminism. Their important feature is that except for providing a basis for formal reasoning, they can also be used for simulation purposes. Petri Nets, already extensively used in the area of manufacturing systems, have one basic weakness: the lack of modularity. As opposed to that, real-time concurrent specification languages are modular and thus allow for hierarchical modelling. This paper demonstrates the use of a real-time concurrent language in modelling of manufacturing systems. The resulting (hierarchical) models are viewed as collections of independent components interacting by synchronous message passing. Such models can be validated by sequential or distributed simulation. The description of component behaviour is separated from the descriptio...

Introduction to the Project Planform is a 26 month long project with a planned starting date of October 1st, 1999. It will be led by Prof Lee McCluskey at University of Huddersfield, Ms Ruth Aylett at the University of Salford, and Dr Maria Fox and Dr Derek Long at the University of Durham. There will be three other post- doctoral members of the team, one at each of the three Universities. The project is funded by the EPSRC, but it also will be supported by the UK National Air Traffic Services Ltd, and CogSys Ltd. The total value of the project is estimated at around 400,000 pounds. Project Relevance Projects such as those sponsored by ARPI [1] and NASA [13] have shown that large-scale planning systems - such as SIPE-II and O-Plan - developed in research centres, can be cost effective. AI Planning systems such as these are domain-independent, that is, their algorithms and representational facilities are logically separate from the model of a particular applicat

Whilst some undergraduate introductions to formal methods play down the role of proof, others have tended to emphasize it as the true payback of using formal methods in the first place. This paper describes how a sports application can be used to illustrate many of these paybacks in a readily understandable way. It illustrates the difficulty of arriving at a formal specification of a complex specification, which is often a collaborative effort between at least two parties, and how this affects the chosen development method. The maintenance of specifications is also considered, in the light of frequent and complex rule changes, in some cases, as a result of "case law" during mid-season. 1 Proof in Formal Methods Many undergraduates are introduced to formal specification, yet learn little or nothing of proof. The emphasis is on the fact that the formal notation provides a precise and unambiguous specification, unlike a natural language specification, or the original requirements docume...

This is an abridged version of chapters 6 and 7 in [Turner and McCluskey 94], which were originally derived from [McCluskey 88]. It introduces the reader to Planning as viewed from an Artificial Intelligence perspective. It also shows how a planning algorithm can be formally specified and then transformed into executable code. Note that the specification was developed without the use of any tools, hence it may contain some syntactic bugs. The planner implementation (and its derivatives) have been used frequently for nearly ten years: in that time no semantic bugs have been found. Hence the application could also be viewed as case study evidence supporting the power of formal techniques. After introducing the reader to planning in Sections 1 and 2, in Section 3 we describe an abstract specification of a planner, In Section 4 we introduce a design level solution to planning in the form of a goal directed algorithm whose basic operation is to achieve goals within a developing, ...