Security Robots vs Guards: the Operational Comparison for Plant Managers 2026

Plant managers and security officers face a calculation in 2026 that no longer goes away. A 24/7 post costs between 15,000 and 25,000 euros per month. A QR-2 costs 3,500 euros. Both numbers are real, both have limits. This article separates tasks cleanly, names trade-offs and provides the basis for the next board paper.

Security Robots vs Guards: the Operational Starting Point

The comparison has five axes: cost per month, detection range, availability, legal suitability, scalability. Anyone who only talks about cost loses the discussion with the works council. Anyone who only talks about detection loses it with the CFO.

Market reality in 2026 shows: the BDSW documents persistent staff shortages in the guard and security industry despite repeated wage increases. Open positions remain unfilled for months. Night-shift turnover is in double-digit percent territory.

The illustrative metric: a 24/7 post mathematically binds 4.8 full-time equivalents. This figure derives from 168 hours of weekly demand, the 38-hour collective agreement week, holiday entitlement under the Manteltarifvertrag, sick leave and shift premiums. Four-point-eight FTE per post is the number that triggers board decisions.

QR-2 runs 24/7 without shift planning, without sick leave, without wage escalation during the contract term. This is not ideology, it is a balance sheet question. The comparison is not either/or. Hybrid models dominate in practice and are treated here as the standard.

Direct Cost Comparison: 24/7 Post Against QR-2

A 24/7 guard post in the DACH region costs between 15,000 and 25,000 euros per month. The range reflects collective bargaining region and qualification level. Premiums for night, Sunday and public holidays come on top. In the West, plant guard posts regularly exceed 22,000 euros per month gross including the service provider margin.

QR-2 as a Robotics-as-a-Service model costs 3,500 euros per month. Fully OpEx, no CapEx, no maintenance invoice, no spare parts inventory. 24-month contract term. Delivery within 48 hours of order. No board-level investment approval required, since below the typical CapEx threshold.

Hidden costs are rarely shown in quotes. These include four to eight weeks of onboarding, turnover cost, training obligations under §34a GewO, uniforms and patrol vehicles. Realistically, these line items add 8 to 12 percent on top of personnel cost. [Source to be added]

The hybrid model QR-2 plus reduced patrol service typically lowers Total Cost of Ownership by 35 to 50 percent compared to pure personnel guarding. [Source to be added] The exact figure depends on perimeter length, number of site entrances and shift model. Anyone who wants a defensible comparison calculation will find the methodology under Guard service cost in detail.

Detection Performance in Field Comparison

Human perception declines measurably after roughly 20 minutes of monotonous observation. [Source to be added] The effect is called vigilance decrement and has been documented in work science for decades. A guard on patrol detects less in the third night hour than in the first. This holds regardless of site size. This is not a question of discipline, it is physiology.

QR-2 combines thermal imaging and person detection. Thermal signatures are reliably detected in darkness, fog and light rain. The detection range for human silhouettes is well above unaided human sight. QR-3 with LiDAR and drone detection supplements this sensor suite for KRITIS perimeters where unmanned aerial vehicles must be expected.

The false alarm rate of autonomous patrol, according to internal Quarero evaluations, lies below that of manual video review in a control room. [Internal study to link or date] The reason is simple: the robot classifies objects before triggering an alarm. A fox is not a person, a moving branch is not an intruder. This pre-classification reduces the load in the control room.

Documentation obligations remain. Every patrol generates audit-proof logs with timestamp, route, detections and alarms. The guard book is not eliminated, it is relieved. For insurance cases and compliance audits, this is an advantage that translates into premiums.

Where Guards Remain Indispensable

There are five tasks a robot cannot or may not take over. This separation is factual, not polemical.

Access control with ID verification and visitor registration requires a person or a certified turnstile with connected identification. Legal responsibility for the admission decision sits with a person.

Escalation management toward intruders benefits from physical presence. The deterrent effect of a patrol in an escalation case is empirically established. A robot documents, a guard intervenes within the legal framework under §34a.

First aid in medical emergencies on the works premises, for instance after a workplace accident on the night shift, is a personnel matter. Defibrillator application, recovery position, initial communication with emergency services.

Legal evidence through witness testimony in court has a different quality than a video log. The two complement each other, neither fully replaces the other.

Key management and lock-up service under VdS guidelines require personalised accountability. Who drew which key when must be attributable by name.

Legal Framework: What Robots May and May Not Do

The standard EN ISO 13482 governs safety requirements for personal care robots including autonomous patrol functions in accessible areas. Quarero platforms are compliant. This compliance is a prerequisite for deployment on works premises with pedestrian traffic.

The EU Machinery Regulation 2023/1230 replaces the previous Machinery Directive from 20 January 2027 and tightens conformity obligations for autonomous systems. Anyone investing in robotics in 2026 should have the manufacturer's declaration of conformity checked against the new regulation.

Data protection: patrol routes and camera operation must be documented in line with GDPR. Concretely this means: record of processing activities and data protection impact assessment for systematic monitoring. The works council must be involved under §87 BetrVG. A camera-equipped patrol on works premises is subject to co-determination.

Robots do not replace a §34a certificate of competence. They supplement it. The control room that verifies robot alarms remains a security trade under the relevant trade law.

NIS-2 obliges essential entities to documented physical protection measures at the perimeter. Article 21 names physical security explicitly. Autonomous patrol with audit-proof logs is a suitable measure for fulfilling this documentation obligation.

Hybrid Model: the Operational Standard Solution

In 2026 the hybrid model is not a transitional solution, it is the standard. Task allocation follows a clear pattern.

Day shift: a reduced guard team for access control, visitor management, key issue, escort of external service providers. These tasks require humans and are well plannable during the day.

Night shift: QR-2 takes over outdoor patrol on defined routes with randomised deviations. Guards are only in the control room, verify alarms and dispatch interventions. The patrol drive across the works premises is eliminated.

Alarm verification by human, detection by machine. This separation uses the strengths of both sides and is superior both in work-science and in economic terms.

Savings potential: typically 8,000 to 12,000 euros per month per site, depending on prior headcount. [Source to be added] The saving arises not through dismissals but through positions not being refilled. In a market with chronic staff shortages, this is the realistic implementation path.

Scaling: additional robot units deliverable in 48 hours. Additional qualified personnel requires three to six months for recruitment, §34a instruction and onboarding. [Source to be added] When expanding sites, this speed differential is operationally decisive.

A concrete implementation outline for an industrial estate with several tenants is available under Perimeter security in industrial parks.

Decision Matrix for Plant Managers

The decision reduces to four site profiles.

Site size above 50,000 square metres of perimeter: QR-2 for 24/7 outdoor perimeter is economical from month one. The patrol distance per shift is physically demanding for personnel and routine for a robot.

KRITIS facility with drone risk: QR-3 with LiDAR is a mandatory candidate. NIS-2 documentation obligations and the specific drone threat at energy, water and telecommunications infrastructure cannot be met with personnel alone.

Indoor logistics hall or data centre: QR-1 covers nightly inspection rounds, checks doors, documents temperature and motion anomalies. The classic plant guard round pattern is automated, personnel dispatches.

Board paper: 24-month OpEx comparison against personnel costs plus a risk markup for staff shortage. The paper calculates both scenarios: pure personnel guarding and hybrid model. The delta figure convinces the CFO. The risk reduction convinces the audit committee.

Pilot phase of 90 days recommended. In this time patrol routes are optimised and alarm thresholds calibrated. The works council agreement is finalised and KPIs are measured. After 90 days data is available to carry a hard decision. The three-tier pricing model provides a corresponding pilot option.

Next Step

Anyone who wants to transfer these numbers to their own site begins with the TCO calculation. The complete methodology with collective bargaining regions, shift models and scaling assumptions is available under Guard service cost in detail. For a 90-day pilot with defined KPIs and works council paper, a short request via Three-tier pricing model is sufficient.