Development of an Integrated Data-Driven Budget Allocation Approach for Maintenance Management in Healthcare Facilities
Bibliographic record
Abstract
Healthcare facilities are fundamental infrastructure assets as their number and quality are common measures of a country’s prosperity and quality of life. Despite Canada being one of the highest countries all over the globe in the health spending, the Canadian hospitals were described in multiple reports as facilities with a crumbling status with a poor condition rating. This was determined according to their high value of deferred maintenance as opposed to their current replacement values. As a result, this study was initiated with the objective of developing a comprehensive asset maintenance and renewals management framework replacing the current approaches in place. The developed framework was thus expected to enhance the performance of hospital buildings assets and efficiently utilize the funds assigned for healthcare facilities on an informed and objective basis. The objective of this study was achieved through four different phases tackling various levels of the healthcare decision-making hierarchy, namely: Asset-Level, Facility-Level and Network-Level. The first pillar introduces an automated priority setting methodology for assets in hospital facilities utilizing multi-criteria decision-making techniques as well as Python-programmed supervised learning algorithms. The following model is concerned with forecasting the possible deterioration in the hospital assets on an integrated mechanism combining between a Matlab-based fuzzy inference system, Markovian models, and metaheuristics. Moving on to a higher level in the decision-making process, a facility renewal scheduling model is advanced to incorporate healthcare-tailored objectives into the planning process. This tri-objective model aims at the reduction of associated wait times and cost while maximizing the performance enhancement gained from the renewal interventions application. Furthermore, unsupervised learning clustering algorithms were used as part of this model to generate a further reduction in the wait times related to renewal intervention applications by grouping relevant interventions together according to the resources available, their location and their priority levels. Finally, a network-level budget allocation model was established relying on the outputs of previous models as inputs for an informed and objective distribution of available budget across hospitals located within one network. The first three models were applied on case study hospitals from Canada and Egypt, and they demonstrated a significant improvement in the current status of assets and facilities. While the final model was applied on a network of hospitals in the province of Alberta and all previous models were re-applied on the current assets and facilities to enable the application of the network-level model. The model results were compared to the actually selected and implemented interventions as well as the allocated budgets, and the model proved an improvement in the prioritization of the assets of 25.97%, a reduction in the deterioration prediction error of 39.29% as compared to the currently implemented mechanism on assets. The facility-level scheduling and clustering models demonstrated a criticality weighted performance enhancement of 28.01%, while maintaining a reduction in the associated downtime of 17.40%. Lastly, applying the network-level budget allocation model resulted in an 33.64% higher network performance for the same budget allocated as per collected records. This proves the capabilities of the developed models in producing more sound and informed decisions relating to healthcare asset management and accordingly, reducing the associated wait times to renewal interventions, refining the overall healthcare performance levels while maintaining minimal budget expenditure possible.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.003 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".