When Smart Systems Optimize the Wrong Thing: Evaluating Security and Sustainability in a Tourist Incident Response System
Mohd. Tabrez Mukadam, Zafar Khan, Mohammed Bilal Shamsi, Mohd. Faizan Shaikh · Journal of Collective Sciences and Sustainability · 2026
Not cached yet. This record was fetched live for this request and is stored after the lookup. Reopening the page should serve the cached copy.
Journal of Collective Sciences and Sustainability · GR Scholastic LLP · article · 2026
Abstract
Smart city emergency response systems routinely fail not because individual sensors malfunction but because designers optimize a single metric, location precision, at the expense of battery life and alert reliability. This failure mode, termed local optimization fallacy, occurs when developers focus too narrowly on one subsystem objective, inadvertently degrading broader system goals such as energy sustainability and network resilience. To address this, we present a 3-Layer System Evaluation Framework that assesses smart infrastructure across technical, sociotechnical, and temporal dimensions.
We validate the framework using a prototype Smart Tourist Safety Monitoring System tested against a simulation dataset comprising 500 tourist trajectories and 2,126 synthetic alert events distributed across 25 geographically realistic urban zones.
Replacing constant GPS polling with adaptive geofencing (Shapely library) and integrating an SMS fallback for temporal resilience, combined with a human-in-the-loop SOS cancellation mechanism, yielded the following simulated outcomes: an 83.5% reduction in GPS polling energy (95% CI [83.4%, 83.5%], paired t test t = -4112.98, p < 0.0001) over a 5-hour tracking session and a 27.7% overall reduction in false-alarm escalations across the simulated zone distribution (35.5% in Safety Zones, 21.6% in Neutral Zones, 3.1% in Danger Zones). All performance results reported are simulation-based estimates obtained using a discrete-event simulator calibrated according to the proposed methodology.
Real-world validation remains an important priority for future work. The findings confirm that in resource-constrained deployments, balanced multilayer optimization consistently outperforms single-metric maximization.
Open the resolved locator · 10.64189/css.26407
Metadata licence: OpenAlex metadata (CC0)