Overview
A global leader in mobility and asset solutions — spanning elevators, escalators, and moving walkways — was operating under a reactive maintenance model that led to costly unplanned downtime, safety risks, and growing customer dissatisfaction. With thousands of units deployed across commercial, residential, and transit infrastructure worldwide, even minor equipment failures could cascade into significant operational disruptions.
The organization partnered with Parkar to build a predictive maintenance platform powered by IoT sensors, Microsoft Azure, and Databricks. The goal was to shift from reactive break-fix cycles to a proactive, data-driven maintenance strategy that could anticipate failures before they occurred, optimize technician dispatch, and ultimately transform uptime into a competitive differentiator.
Challenge
The existing maintenance approach was unsustainable as the organization scaled its global footprint. Key challenges included:
- Unplanned Downtime: Equipment failures were detected only after they occurred, leading to extended outages, emergency repair costs, and negative impacts on building operations and tenant experience.
- Safety and Compliance Risks: Aging assets and inconsistent maintenance schedules introduced safety hazards and made it difficult to meet evolving regulatory and compliance standards across different regions.
- Escalating Maintenance Costs: A reactive model meant higher spare-parts inventory, frequent emergency dispatches, and inefficient technician utilization — all driving up total cost of ownership for both the organization and its customers.
Solution
Parkar designed and delivered an end-to-end predictive maintenance platform built on four core components, enabling the organization to harness real-time data from thousands of connected assets worldwide.
IoT Data Backbone
Thousands of IoT sensors were deployed across elevators, escalators, and walkways to capture real-time telemetry — including vibration, temperature, motor current, door cycle counts, and ride quality metrics. Data was ingested through Azure IoT Hub, providing a reliable, scalable pipeline from edge devices to the cloud.
Azure + Databricks Analytics Engine
A Lakehouse architecture built on Azure Data Lake Storage and Databricks provided the foundation for advanced analytics. Machine learning models were trained on historical maintenance records, failure logs, and sensor data to predict component degradation and remaining useful life. These models enabled condition-based maintenance scheduling, replacing rigid time-based intervals with intelligent, data-driven triggers.
Governed Data Fabric
Microsoft Purview and Databricks Unity Catalog were implemented to create a governed data fabric across the organization. This ensured data lineage, quality, and access controls were maintained as the platform scaled — critical for regulatory compliance and cross-regional data sharing.
Decision Intelligence Layer
Interactive dashboards and AIONIQ copilots were deployed to put actionable insights directly into the hands of maintenance managers, field technicians, and executive stakeholders. Technicians received prioritized work orders with predicted failure windows, while leadership gained visibility into fleet health, cost trends, and SLA performance across regions.
Key Results
- 90% Reduction in Unplanned Downtime: Predictive models identified potential failures days or weeks in advance, enabling proactive intervention and dramatically reducing emergency service calls.
- 20% Lower Maintenance Costs: Condition-based scheduling optimized technician routes, reduced unnecessary inspections, and minimized spare-parts waste, delivering significant cost savings across the fleet.
- Improved Safety and Compliance: Continuous monitoring and automated alerting ensured assets met safety thresholds and regulatory requirements, reducing incident risk and audit exposure.
- New Revenue Stream — Uptime-as-a-Service: The platform enabled the organization to offer premium uptime guarantees and predictive maintenance contracts to building owners and operators, creating a new, recurring revenue model.
- Higher Customer Satisfaction: Tenants and building managers experienced fewer disruptions, faster issue resolution, and greater transparency into equipment health, driving improved Net Promoter Scores and contract renewals.
The Future
- Smart Buildings Integration: Expanding IoT connectivity to integrate elevator and escalator data with broader building management systems for holistic facility intelligence.
- Energy Efficiency Optimization: Leveraging sensor data and ML models to optimize energy consumption across mobility assets, supporting sustainability goals and reducing operational costs.
- Next-Gen Asset Management: Evolving from predictive maintenance to prescriptive maintenance, where the platform not only predicts failures but recommends optimal repair strategies, parts sourcing, and technician assignments in real time.
"Our platform's strength lies in its adaptability — combining IoT, cloud analytics, and AI to turn raw equipment data into real-time decisions that keep people moving safely and efficiently."