Service Robot Elevator Integration: How Robots Travel Between Floors in Hotels and Hospitals

The most common reason a hotel or hospital decides not to deploy a service robot is also the most commonly misunderstood: the elevator. The board approves the budget, the operations team is excited, the supplier demos the unit on a single floor — and then someone asks, "But how does it get to the guest room on the 12th floor?" If the answer is "we'll figure that out at install time," the project is already at risk.

Multi-floor delivery is exactly the use case that justifies a service robot in the first place. A single-floor restaurant or ground-floor clinic does not have a strong economic case for a $3,000-5,000 per unit robot. The economic case is built on cross-floor trips: room service in a 200-room hotel, medication and lab samples in a multi-story hospital, mail and parcels in a 30-floor office tower. The robot is the answer to those use cases — and the elevator is the bottleneck that determines whether the answer works at all^[1].

This guide gives you the framework we use at 云南智创机器人（YNZC） for elevator integration projects in Vietnam, Thailand, Singapore, Malaysia, Indonesia, and the Philippines. It covers the three integration approaches, a 5-step technical walkthrough, common failures we have learned to avoid, and a 10-question checklist to take to your robot supplier before signing a contract.

1. Why Multi-Floor Delivery Is the Use Case — Not a Nice-to-Have

Most published case studies about service robots in hospitality focus on ground-floor or single-floor deployments. The headline metrics — labor hours saved, table turnover, customer satisfaction — are real but they are not the high-value case. The high-value case is multi-floor, because that is where the labor pain is concentrated and where the cost of a service robot is easiest to justify.

Consider a mid-size hotel in Bangkok or Kuala Lumpur. The property has 180-250 rooms spread across 15-25 floors. Room service generates 60-120 delivery trips per day. The runner team needs 2-4 staff on shift during peak hours, and they spend a significant portion of their time waiting for the elevator. According to a 2024 STR report, room-service runner labor accounts for 12-18 percent of total room division payroll in mid-scale Asian hotels^[2].

Now consider a 400-bed hospital in Jakarta or Ho Chi Minh City. Nurses spend an estimated 30-45 minutes per shift walking between the pharmacy, the central supply, the laboratory, and the wards. The WHO reports that nurses in Southeast Asian tertiary hospitals spend 25-40 percent of shift time on logistics tasks rather than direct patient care^[3]. A service robot that can take the elevator autonomously directly addresses both the labor cost and the patient care quality problem.

The math is straightforward. A service robot for hotel room service is typically priced in the range of around $3,000-5,000 per unit, with a payback period of 12-18 months in most Southeast Asian markets. Hospital logistics robots are similar in price but have a longer useful life and a higher labor displacement factor, so payback is often 10-14 months. None of these numbers work, however, if the robot can only operate on a single floor.

2. The Three Main Approaches to Elevator Integration

Across our deployment database covering 240+ commercial installations, elevator integration falls into one of three categories. Each has different cost, timeline, and reliability characteristics, and each is appropriate for a different building profile.

2.1 Approach A: Direct API Integration

The cleanest approach. Newer elevator controllers from KONE, Mitsubishi, Otis, Schindler, Hitachi, and Hyundai expose documented APIs that allow an external system to call the elevator, select a floor, and query its status. If your building has such a controller — typically installed in the last 8-10 years — a direct API integration is the lowest-friction path: a 1-2 day install, low hardware cost, and high reliability because the robot and the elevator share the same source of truth for floor status.

2.3 Approach C: Cloud-Mediated Vendor API

The third path is increasingly common. Major elevator vendors now offer their own cloud services (KONE 24/7, Otis ONE, Schindler Ahead) that can dispatch and control the elevator. The robot supplier integrates with the cloud API rather than the on-site controller. This approach is the most future-proof — vendor upgrades are automatically inherited — but it requires an active subscription with the elevator vendor and adds a monthly fee per elevator.

2.4 Comparison Table: Which Approach When?

3. How the Integration Works: A 5-Step Technical Walkthrough

Regardless of which approach is used, the end-to-end flow is the same. The robot, the elevator, and the building network cooperate to complete a single multi-floor trip. The following walkthrough uses a hotel delivery scenario from our Bangkok deployment database.

3.1 Step 1: Trip Request from the Front-of-House System

A guest calls room service for two bottles of water. The order is entered into the hotel PMS, which routes it to the service robot's dispatch queue. The dispatch queue assigns the trip to a specific robot on the floor closest to the kitchen — typically the ground or basement floor. The dispatch message includes pickup, drop-off, and any special handling instructions (e.g., "guest requested sparkling water," "no contact delivery").

3.2 Step 2: Path Planning and Elevator Request

The robot receives the trip and calculates a path from the kitchen to the elevator bank. The robot's fleet management system sends an elevator call request to the elevator controller via the API or the IoT bridge. The request specifies the robot's current floor, the destination floor, and the desired direction (up or down). Most modern elevator controllers treat robot requests with the same priority as guest calls; some buildings configure a "service priority" that holds the elevator longer for robot trips.

3.3 Step 3: Robot Enters the Elevator

The elevator arrives at the robot's floor and the doors open. The robot uses its onboard LiDAR and 3D cameras to detect the elevator car position, the door width, and the height of the gap between the cabin floor and the lobby floor. The robot enters the cabin, positions itself in a designated corner (typically the rear-left or rear-right to avoid blocking human passengers), and signals that it is "on board."

3.4 Step 4: Floor Selection and Transit

The floor selection happens in one of two ways. In Approach A (direct API), the robot's fleet management system sends the destination floor directly to the elevator controller. In Approach B (IoT bridge), the bridge emulates a button press on the elevator's floor panel. In both cases, the elevator moves to the destination floor while the robot remains stationary in the corner.

3.5 Step 5: Exit and Final Delivery

When the elevator arrives at the destination floor and the doors open, the robot exits, the doors close, and the robot continues to the guest room or the hospital ward. The robot sends a delivery-completed signal back to the dispatch system, which logs the trip duration and updates the room-service dashboard. The end-to-end time for a typical 10-floor round trip is 4-7 minutes, of which 60-90 seconds is spent inside the elevator.

4. Real-World Use Cases: Hotel, Hospital, and Office

The integration pattern is identical across industries, but the operational priorities differ. A hotel cares about guest perception. A hospital cares about hygiene and reliability. An office cares about throughput.

4.1 Hotel Room Service

Hotels deploy multi-floor service robots for room service deliveries (water, towels, toiletries, food), amenity deliveries (extra pillows, ice), and concierge items (documents, forgotten items from the front desk). The economic case is strongest in 3-4 star properties with 150-400 rooms. In our Bangkok and Kuala Lumpur deployments, multi-floor robots have replaced 1.5-2.5 FTE runner positions per shift, with payback in 12-18 months. Guest satisfaction is typically neutral to positive — the contactless delivery option is increasingly popular, and most properties report that guest scores for "staff helpfulness" do not decline when robots handle a portion of the trips.

4.2 Hospital Logistics

Hospitals use multi-floor robots for pharmacy-to-ward medication delivery, central supply delivery, lab specimen transport, meal tray delivery, and document delivery. The economic case is driven less by direct labor displacement and more by nurse time recovery. A typical 300-bed hospital in our deployment database recovers 8-12 nurse hours per day — time that goes back to direct patient care. Reliability is the dominant requirement: hospital robots typically target 99 percent on-time delivery and 100 percent audit trail for medication deliveries^[5].

4.3 Office and Commercial Buildings

Office deployments focus on mail and parcel delivery between floors, food delivery from the ground-floor cafe to upper-floor tenants, and document transport. The economic case is more variable and depends heavily on the building's existing mailroom staffing. Multi-tenant office buildings of 20+ floors in Singapore, Jakarta, and Bangkok have started pilot deployments, typically with one or two robots per building, focused on peak-hour lunch delivery to reduce the load on the food court elevators.

5. Common Failures and How to Avoid Them

Across 240+ deployments, we have seen four recurring failure modes. None of them are fundamental technical issues; all of them are upstream planning gaps.

6. What to Ask Your Robot Supplier About Elevator Integration

Before signing a contract, take this checklist to the supplier. A confident supplier will have ready answers; a weak one will hedge. We have seen the difference between a 4-week deployment and a 4-month deployment come down to whether these questions are asked up front.

7. The Multi-Floor Future: What's Next After Elevator Integration

Elevator integration is the foundation for the next generation of service robot deployments. Once a robot can reliably take the elevator, the operating model shifts from "one robot per floor" to "fewer robots serving more floors," which changes the unit economics. A 2025 McKinsey analysis estimated that multi-floor capable robots are 35-50 percent more labor-efficient than single-floor deployments in hotels with more than 150 rooms^[7].

The next frontier is integration with automated parking systems, sliding doors, security turnstiles, and fire alarm systems. The same IoT bridge pattern that allows a robot to call an elevator also allows it to badge through a security gate, open a fire door during an emergency, or signal a parking gate to lift. Vendors that have done elevator integration well are typically the same vendors that can extend to these adjacent integrations quickly, which is another reason to choose an experienced supplier.

If you are evaluating a service robot for a multi-floor hotel, hospital, or office, start with the elevator conversation. It will tell you more about the supplier's deployment depth than any demo.

About the Author

YNZC Editorial Team — 云南智创机器人（YNZC） marketing engineering group. 8+ years deploying service robots across Vietnam, Thailand, Singapore, Malaysia, Indonesia, and the Philippines. Reviewed by Jiang Hailong (Founder, 10+ years in commercial robotics). About our team →