Overnight Mall Patrol Robot Reduces Security Costs by 62% and Detects 31 Incidents in Year One

A regional shopping mall operator managing a 180,000 m² mixed-use retail complex in southern China faced a persistent challenge with overnight security operations. The mall — comprising five retail floors, two basement car parks, a food court, a cinema, and a service corridor network — required 22 overnight security guards to maintain acceptable coverage from 22:00 to 08:00. The annual cost of the overnight security operation represented 34% of the mall’s total facilities management budget.

Despite the significant staffing investment, the operations director identified three structural weaknesses in the overnight security model. First, coverage consistency: with 22 guards covering 180,000 m², individual patrol routes were completed only once every 90–120 minutes — leaving large sections of the mall unmonitored for extended periods. Second, human factors: incident log analysis showed that guard response times degraded significantly between 02:00 and 05:00, with average response times to reported incidents extending to 14 minutes during this window. Third, facility risk blind spots: the overnight team had no systematic process for detecting non-security hazards such as water leaks, gas accumulation, or electrical overheating — risks that had caused two significant insurance claims in the previous three years.

The operations director sought a solution that could maintain consistent patrol coverage throughout the overnight period, provide real-time incident detection and alerting, and extend monitoring to facility hazards beyond traditional security events — while substantially reducing the labour cost of the overnight operation.

Shaanxi Smart Innovation Future Technology’s solution team conducted a two-week site survey, mapping patrol routes across all five retail floors, both basement car parks, and the service corridor network. Six RZTL-1W wheeled-legged robots were specified — one per retail floor and one shared between the two basement levels — with autonomous charging docks installed at central locations on each floor to enable continuous overnight operation without manual battery management.

The RZTL-1W was selected for its ultra-quiet wheeled locomotion mode — critical for overnight operation in a retail environment where noise from a legged robot would be unacceptable — combined with the leg-based obstacle clearance capability needed to navigate retail fixtures, service equipment, and the varied terrain of the basement car parks. Each robot was equipped with a dual-spectrum pan-tilt camera (visible light + thermal imaging), a gas detection sensor array (CO, LPG, and smoke), a water leak detection sensor, and a two-way audio module.

The cloud management platform was configured with the mall’s existing CCTV and fire alarm systems, enabling the robot fleet to serve as a mobile sensor layer that complemented the fixed infrastructure. AI behaviour analysis was configured to detect intrusion, loitering, and physical altercations; thermal imaging was set to alert on temperature anomalies in electrical distribution areas and kitchen exhaust systems; and the gas sensor array was configured to trigger immediate evacuation protocol alerts if readings exceeded safe thresholds.

In the first year of operation, the autonomous patrol system detected 31 incidents across all categories: 19 security events (including 11 after-hours intrusion attempts, 5 vehicle break-ins in the basement car parks, and 3 cases of vandalism), 10 facility anomalies (including 7 water leak detections and 3 electrical overheating alerts), and 2 critical safety events that the operations team identified as the system’s most significant contributions.

The two critical safety events were a gas accumulation alert in the food court kitchen area at 03:15 on a Tuesday morning, and a thermal anomaly in an electrical distribution room on the third floor that was subsequently identified as a failing transformer. In both cases, the robot’s sensor systems detected the developing hazard before it reached a level that would have triggered the building’s fixed detection infrastructure — providing an estimated 45–90 minutes of additional warning time. The mall’s insurance provider reviewed the incident reports and reduced the facility’s annual premium by 12% in recognition of the improved risk management capability.

The overnight security team was reduced from 22 guards to 8 — with the remaining staff managing the cloud platform, responding to robot-generated alerts, and handling situations requiring physical intervention. Security labour costs fell by 62%, with the full system investment recovering in 11 months based on labour savings alone. The operations director subsequently extended the robot patrol schedule to include early morning hours (06:00–08:00) and weekend daytime periods, further expanding the system’s coverage and value.