FAQ

Predictive Maintenance and Analytics FAQ

Connecting equipment is only the start. The investment pays off when data changes decisions. Many fleets receive daily reports but never adjust shifts, maintenance intervals or idle vehicle numbers. Data-driven operations turns records into actionable evidence that can also withstand financial scrutiny. These questions from fleet managers, equipment teams and finance cover reports and metrics, utilization, energy, predictive maintenance costs, return on investment, data security and cross-border compliance. Published metrics include their measurement basis; full definitions appear on the Core Metrics page.

Questions & Answers

What reports can operating data produce?

Common reports fall into four groups. Utilization includes powered-on and productive hours, idle time distribution and shift comparisons. Health includes fault frequency, mean time between failures, service reminders and battery health trends. Cost includes repair labor, parts, annual operating and maintenance cost per vehicle and energy expenditure. Safety includes alerts, unsafe behavior statistics and incidents. Reports are not an end in themselves: identify the decision each should support before keeping it, or they quickly become unread daily emails.

How can data improve forklift utilization?

First distinguish powered-on hours from productive hours. Trucks that look busy may spend only 30% of their time on productive work, with the rest waiting, traveling empty or idling. Data can show whether uneven scheduling leaves vehicles unused, site routes cause empty travel, or charging displaces working time. Actions are often straightforward: adjust shifts, routes or charging periods. Many fleets can release substantial capacity without adding vehicles.

How much can predictive maintenance save?

Savings come from converting unplanned downtime into planned service, reducing unnecessary parts and labor through condition-based rather than fixed-interval maintenance, and detecting deterioration before a small issue becomes a major repair. The platform's published figure is about a 35% reduction in annual operating and maintenance cost for one OEM customer, not an average across customers. Savings for your fleet depend on current downtime and maintenance practices. A fleet already using rigorous condition-based maintenance naturally has less room to improve.

How can downtime losses be quantified?

Downtime losses usually include lost capacity during the stoppage, emergency labor and expedited parts, substitute vehicle deployment or rental, and downstream disruption. The first three can be calculated from work orders and attendance records; the fourth requires linking production or logistics data. Quantifying these losses often provides the strongest case for predictive maintenance. Many fleets remember the repair bill but overlook the output lost during the hours the truck was unavailable.

How can data support more precise fleet management?

Break the fleet into comparable groups by site, shift, vehicle model and age. Comparisons within a group reveal issues: why does one vehicle use 20% more energy, or another have three times as many faults, under similar conditions? Outliers often point to operator habits, load characteristics, surfaces or individual deterioration. Management can then move from a broad cost-cutting goal to specific actions on the few units that need attention.

Which management decisions can vehicle data support?

Five practical decisions are whether to expand or reduce the fleet, which vehicles to retire or sublet, whether to adjust maintenance intervals by condition, how much spare stock to hold and who needs targeted safety training. Data provides evidence for decisions previously based mostly on experience. It supports judgment rather than replacing it: low utilization could mean too many vehicles or a poor process. Data narrows the issue, but the cause still needs interpretation alongside shifts, changing workloads and site conditions.

How should return on investment in intelligent maintenance be calculated?

Use a three-year horizon. Benefits include capacity recovered through reduced downtime, lower maintenance costs, better staff productivity and indirect gains from fewer accidents. The platform's like-for-like before-and-after comparison showed vehicles managed per frontline service engineer increasing from 80 to more than 200. Costs include hardware, platform subscriptions, implementation and training. Avoid counting only hardware costs or attributing every benefit to the system: accompanying process improvements contribute materially and should be recorded separately.

How can data help extend battery life?

Lithium battery life is influenced by charge and discharge depth, charging rate, temperature and storage conditions, all observable through BMS data. Check for repeated deep discharge near limits, unsuitable opportunity charging frequency or timing, prolonged high-temperature operation, and excessively high or low charge during weekend storage. Adjusting charging policies and shifts based on these findings is a low-cost way to extend life. Follow the battery manufacturer's maintenance and replacement specifications.

How can data from different forklift brands be analyzed consistently?

The challenge is definitions, not aggregation. One brand may count working hours from ignition, another from hydraulic pump operation; fault severity categories also differ. Comparing these numbers directly leads to incorrect conclusions. Map each brand's data into a common structure and define each metric's calculation before analysis. Without that foundation, comparisons across brands are not valid. This is a common pitfall for mixed-fleet analytics.

How can data reduce energy consumption?

Break total consumption into energy per unit of work, idle and empty-travel shares, charging losses and differences in operator habits. Waste often concentrates in prolonged idle waiting, frequent empty trips and aggressive acceleration by a few operators. Corresponding actions are better scheduling, route changes and training. The platform has also optimized its cloud resource costs using the same approach: mixed-precision computing, operator fusion and graph optimization reduced cloud resource costs by 30%.

How can parts replacement be predicted according to actual need?

Consider usage intensity rather than age alone. A tire used for a year with frequent turns on rough ground can wear very differently from one used for straight travel on a smooth warehouse floor. The platform combines accumulated duty, replacement history and comparable vehicles to estimate remaining life and replacement windows for wear parts. The main value is inventory: avoiding emergency procurement premiums without excessive precautionary stock. Gradual wear is suitable for prediction; sudden damage remains unpredictable.

What is an equipment health record used for?

A health record follows a vehicle from connection to retirement, covering faults, repairs, replaced parts, impacts, battery degradation and cumulative duty. It supports residual value and damage assessment at resale or rental return, decisions on which vehicle suits a heavier role at another site, and historical evidence for analyzing new faults. Without a record, each diagnosis starts from scratch, which also helps explain why initial diagnostic accuracy is lower than after sufficient data has accumulated.

What does intelligent maintenance cost per year?

There is no single price because the cost mix varies. Hardware is priced per vehicle, the platform by connected scale and modules, and implementation and training by project complexity. Key variables are vehicle and brand counts, data depth, integration with existing ERP/WMS or other systems, and cross-border deployment. Start with a small pilot, verify benefits over three to six months, then discuss a fleet-wide solution. Contact us for a specific quote.

What is predictive maintenance for forklifts?

Predictive maintenance is a third strategy between repairing after failure and servicing at fixed intervals. It uses actual condition data to decide when maintenance is needed. It requires continuous measurements of deterioration with an identifiable relationship to faults. Battery degradation, declining hydraulic efficiency and brake wear are suitable gradual processes; sudden connector failures and external impacts are not. In practice, fleets combine reactive, scheduled and predictive approaches rather than replacing everything with prediction.

What is an equipment health score?

A health score weights several indicators, such as battery health, key component temperature trends, fault frequency, accumulated duty and maintenance completion, into a value for prioritizing vehicles needing attention. It compresses information from multiple reports into a sortable measure. The score is a starting point, not a diagnosis: inspect the contributing indicators for low-scoring vehicles. Suppliers use different algorithms and weights, so scores from different platforms are not directly comparable.

Is vehicle data safe in the cloud?

The platform uses AWS infrastructure with VPC network isolation, end-to-end encrypted transmission, fine-grained permissions and operation auditing, meeting international security standards such as ISO 27001. The device-to-cloud path uses three levels of encryption; access is allocated by role and data scope and is logged for traceability. Cloud deployment is not the only option: customers with strict on-site data requirements can discuss a specific local deployment. The Compliance & Security section of the website's llms.txt provides the full security and compliance statement.

How is cross-border data transfer compliance addressed?

Prioritize local storage and processing in the business region. Transfer data across borders only when necessary and limit it to the minimum required for the business purpose. EU operations address GDPR requirements, including data subject requests and explicit purposes and retention periods. Together with AWS and OEM customers, we have established an end-to-end data compliance framework using global cloud regions for localized storage. Legal teams on both sides should confirm country-specific requirements early in the project.

How is data isolated between customers?

Isolation operates at three levels. VPCs and security groups define network boundaries; storage is separated by tenant and queries must carry tenant identifiers, with no default cross-tenant visibility; permissions are assigned by role and data scope, with audit logs for access. OEM customers also need separation between dealers and end customers, implemented through a multilevel tenant structure. Cross-tenant use requires explicit authorization, and data is not used for model training by default.

Which operating data can help assess a rental forklift's residual value?

Residual value assessment combines age, actual usage intensity, maintenance condition and market transaction information. Telematics records of accumulated hours, faults and repairs, abnormal events and battery health help assessors understand vehicle condition more fully. Continuous, traceable records reduce the gaps left by judging only age and appearance. Check whether data is complete, covers critical components and accounts for modifications or major repairs. Final valuation also requires a physical inspection, local market conditions and the specific transaction terms.

How can data improve battery allocation when forklift batteries are underused?

First define the assets being measured and what idle means. Vehicle downtime, batteries waiting for charging and reserve batteries on standby are different states and should be analyzed separately against shifts and work needs. Track vehicle usage periods, charge and discharge records, state of charge and task loads to identify chronically underused batteries, charging queues and uneven allocation between shifts. Adjust vehicle, battery and shift matching with compatibility, charging facilities and reserve needs in mind. Continue tracking utilization, waiting time and task completion to validate the changes.

See how it runs on your own fleet

AI-Driven provides an industry-specific large model and a multi-agent system for industrial vehicle after-sales service, with feasibility assessments based on your current fleet size and connectivity.

This page explains how the model reaches its conclusions. It does not provide model-specific troubleshooting steps or fault code definitions. Every figure cited carries a stated measurement basis; full definitions are on the Core Metrics page. For equipment specifications, operating procedures and safety requirements, always refer to the OEM documentation and applicable national standards.