Graham Winstanley

AI planning technology as a component of computerised clinical practice guidelines

The UK National Health Service (NHS) is currently undergoing an intensive review into the way patient care is designed, delivered and recorded. One important element of this is the development of care pathways (clinical guidelines) that provide a reasoned plan of care for each patient journey, based on locally-agreed, evidence-based best practice. The ability to generate, critique, and continually evaluate and modify plans of patient care is considered important and challenging, but in the case of computerised systems, the possibilities are exciting. In this paper we outline the case for incorporating AI Planning technology in the generation, evaluation and manipulation of care plans. We demonstrate that an integrative approach to its adoption in the clinical guideline domain is called for. The PROforma Clinical Guideline Modelling Language is used to demonstrate the issues involved.

DISTRIBUTED AND DEVOLVED WORK ALLOCATION PLANNING

Work allocation planning is a vital and notoriously difficult task in areas characterized by large work forces, contiguous and sometimes overlapping shifts, regulatory and corporate constraints, and tightly managed work force establishments. In the allocation of nurses to ward shifts, forward planning is widely acknowledged to be essential, but the majority of efforts in this highly constrained area rely on manual techniques. Health service policy is being driven increasingly towards maximum efficiency and accountability, resulting in the need for more accurate long-term planning. This paper discusses an approach to this problem, using a strategy of distributing and devolving the computational effort required. In the pre-processing stage, the staff to be rostered are treated as semi-autonomous agents, each with individual responsibility for their initial assignment, and communication with a global constraint solving (CLP) agent. This has proved to be intuitive to build and effective in use.

Activity aggregation in model-based AI planning systems

When model-based planning systems are scaled up to deal with full-sized industrial projects, the resulting complexity in the project-specific model and production plan can create serious problems, not only in dealing with such complexity computationally, but also in user-acceptance. In the model-based planning system described in this paper, activities are dynamically generated, inherently at the detailed level of individual physical components. However, it is possible to intelligently group together collections of components which would be common to realistic work packages, and hence schedule on the basis of virtual components existing within an abstraction hierarchy. This paper describes a technique of project planning within an integrated design/planning system, which exploits fundamental knowledge of engineered systems and provides powerful and flexible planning functionality.

The impact of design decisions on post occupancy processes in school buildings

Purpose – This paper aims to review the impact of design and specification decisions for major works during post‐occupancy processes; the routine maintenance and management of school premises. It also seeks to explore the relationship between the main stakeholders and how this impacts decision‐making and the post‐occupancy operation of school buildings.Design/methodology/approach – In addition to a literature review, qualitative data were also obtained through a focus group; a steering group. The steering group consisted of stakeholders tasked with delivering and managing school premises in the Sussex County.Findings – The findings contribute to understanding the impact of design decisions on post‐occupancy processes in schools. A basic model is also presented as a guide for requirements and decision mapping in post‐occupancy design decision processes in schools.Research limitations/implications – The study was conducted with the participation of a representative sample of stakeholders. There may be the n...

Adapting Bottom-up, Emergent Behaviour for Character-Based AI in Games

It is widely acknowledged that there is a demand for alternatives to handcrafted character behaviour in interactive entertainment/video games. This paper investigates a simple agent architecture inspired by the thought experiment “Vehicles: Experiments in Synthetic Psychology” by the cyberneticist and neuroscientist Valentino Braitenberg [1]. It also shows how architectures based on the core principles of bottom-up, sensory driven behaviour controllers can demonstrate emergent behaviour and increase the believability of virtual agents, in particular for application in games.

Facilitating organisational learning through causal mapping techniques in IS/IT project risk management

Information System and Information Technology (IS/IT) development and implementation have become more difficult with the rapid introduction of new technology and the increasing complexity of the marketplace. IS/IT projects often encounter a range of problems that can be described as failure. Thus, learning from an analysis of past projects and from the issues contributing to failure is becoming a major stage in the risk management process. In IS/IT projects, it is common for groups of stakeholders to participate in planning and management. One important element in these activities is risk assessment, that is, the identification of potential risks and their interrelationships throughout the project lifecycle. The ability to visualise cause and effect risk networks and the capability for interactive network building and modification have the potential for individual and group risk identification, justification and prediction. In this paper we introduce Causal Mapping as a method of accomplishing this, and describe two experiments: one carried out with a group of masters-level students and a second with practitioners from a government organization who had experienced an IS/IT project failure. These two exploratory experiments have demonstrated the potential (and also some of the problems) of the approach in identifying problem areas in past projects, through the collaborative construction of cause and effect maps that allow project participants to visualise their perceptions.

A co-creation platform for post-occupancy decision support

Purpose – The purpose of this paper is to report on partial findings of a research project on post‐occupancy design in school projects. It provides a brief review of key factors, and major players in post‐occupancy processes.Design/methodology/approach – Subsequent to literature review for the research, the primary data was obtained from a steering group (stakeholder forum) comprising of a sample of primary and secondary school representatives, local authorities and building professionals. Following the data collection, visits were arranged to the selected schools, and local councils and semi‐structured interviews with school and council authorities were carried out. At the same time, research meeting and directed studies were held to best explore the applications for devising the toolkit. Simulated and real cases for making decisions in post‐occupancy design were selected to devise a model‐base to test out the applicability of the devised prototype of the toolkit.Findings – This study provided clarificat...

A hybrid AI approach to staff scheduling

Assigning staff to specific duties according to their contract, qualifications, skills, etc. within a working environment characterised by multi-disciplinarity and statutory regulations is problematic. This paper discusses an approach to nurse rostering, using a strategy of distributing the computational effort required in the scheduling process. The technique involves a hybrid approach that devolves responsibility for different aspects of the problem. In the pre-processing stage, the staff to be rostered are treated as semi-autonomous agents, each equipped with heuristics to guide their initial assignment. Compilation of individual rosters is followed by a scheduling phase in which a constraint solving agent applies constraint logic programming (CLP) techniques in the generation of ‘acceptable’ rosters.

A computer-based configuration and planning system

Abstract Product-oriented domains are characterized by their commitment to one product, or a relatively small range of similar products. In many cases, their products are large, highly complex and costly, and, because of the reliance of such manufacturing companies on a steady, but sometimes small, flow of manufacturing projects, the accurate, cost- effective and timely production of bids to prospective clients is crucial to their survival. The paper outlines the results of research into the application of expert-systems techniques to the modelling, configuration and planning of product-oriented manufacture, and, in particular, to one particular industry: the manufacture of flight simulators.

A Temporal Pattern Predictor for Virtual Characters

This paper discusses a prototype of a temporal pattern predictor, which was built on specifications derived from the descriptions of the “Ergotrix” temporal memory network in Valentino Braitenberg’s “Vehicles” (Braitenberg, 1984). The prototype was developed as a component for a control architecture for virtual characters.