Work

DSSOU: A Design Science Framework for NHS Non-Clinical Space Utilisation Optimisation Integrating Time-Series Forecasting, Machine Learning, and Linear Programming

Year: 2026
Type: article
Abstract: Non-clinical bookable units-meeting rooms, hot-desks, and carparking spaces-represent a significant but analytically underexamined dimension of NHS facility management.While predictive analytics and optimisation methods are mature in the clinical capacity literature, no published framework integrates these methods specifically for NHS non-clinical space utilisation.This paper proposes DSSOU (Decision-Support System for Occupancy and Utilisation), a four-component design-science artefact grounded in the Hevner et al. design-science research methodology.DSSOU integrates: (i) temporal pattern analysis using ARIMA and Multi-Seasonal ARIMA; (ii) contextual demand prediction using Random Forest and Gradient Boosting; (iii) prescriptive space allocation using Linear Programming; and(iv) an interactive decision-support dashboard.The framework is designed to operate on NHS booking system data spanning 2016-2024 and is specified at sufficient formal detail to support immediate implementation.The data model, the formal optimisation formulation, the prediction model specification, and the artefact evaluation protocol are presented, together with a candid assessment of which components are directly supported by established precedent and which constitute genuinely novel integration.DSSOU is presented as a conceptual design-science proposal: no experimental results are reported, and a detailed, falsifiable evaluation protocol is specified by which the framework's central claims should be tested.This paper constitutes the methodological proposal stage of a planned empirical study.
Language: English
Open Access status: diamond
OpenAlex ID: W7203833197
FWCI:
Cites: 9
Cited by:
Related to: 10