Case Study

Customer Success Story

Transportation & Fleet Operations · AWS Generative AI

Nuevas Builds a GenAI-Powered Predictive Maintenance Platform on AWS to Plan Fleet Maintenance Before Breakdowns Occur

Indovision Services helped Nuevas build a predictive maintenance platform on AWS that streams OBD-II / CAN bus telemetry from commercial fleet vehicles, detects developing faults with rules and machine-learning anomaly scoring, explains each finding in plain language with Amazon Bedrock and drafts maintenance work orders, with dispatchers and maintenance staff still deciding every repair, schedule and vehicle restriction.

CAN !Potential faultUnit 214 · coolant temp Draft work orderHigh Evidence: DTC P0217, temp trendCheck: thermostat, coolant levelRef: approved service manual §4.2 ApproveEdit Explained by Amazon Bedrock People approveevery repair
EarlierTarget time to detect a potential vehicle fault (decrease)
FewerTarget unplanned breakdowns (decrease)
LongerTarget maintenance planning lead time (increase)
LessTarget effort to prepare each work order (decrease)
Overview

One view of vehicle health, from raw telemetry to a reviewed work order

Nuevas Technologies runs a commercial vehicle fleet whose dispatchers, maintenance planners and workshop teams depend on knowing which vehicles are healthy and which need attention. Vehicle gateways report timestamped OBD-II / CAN bus data (diagnostic trouble codes, engine temperature, battery signals, vibration where available, mileage and GPS), and the platform serves dispatchers, maintenance planners, workshop staff and platform administrators through an existing fleet-management system or an optional operations portal.

To move from reactive repairs to planned maintenance without handing control to automation, Nuevas Technologies turned to Indovision Services, an AWS partner focused on AI/ML platforms, to design and build a GenAI-powered predictive maintenance platform on AWS. Telemetry is ingested through AWS IoT Core and Amazon Kinesis Data Streams, rules and Amazon SageMaker AI models detect developing faults, and Amazon Bedrock turns each verified health event into a plain-language explanation and a draft work order, grounded in the evidence and in approved service manuals and always reviewed by people. The design is secured from day one, treats lost vehicle connectivity as normal, and is costed per vehicle so spend stays visible as the fleet grows.

At a glance

  • CustomerNuevas Technologies
  • IndustryTransportation & fleet operations
  • PartnerIndovision Services
  • Core serviceAmazon Bedrock, with AWS IoT Core and Amazon SageMaker AI
  • UsersDispatchers, maintenance planners, workshop staff and platform administrators
The Challenge

Faults surfaced late, in a form that was hard to act on

Maintenance was largely reactive. The information that could have warned the team earlier was technical, scattered and slow to turn into action.

Late fault discovery

Faults were noticed only after performance deteriorated or a breakdown occurred, causing roadside breakdowns, route disruption and unplanned workshop visits.

Hard-to-read diagnostics

Raw diagnostic codes and telemetry did not explain operational priorities, so dispatchers and planners spent time interpreting signals or risked missing them.

Disconnected context

Vehicle health data, route plans and workshop availability were considered separately, making planning reactive and prioritization inconsistent.

Manual searching and drafting

Workshop teams searched service manuals and prepared work orders by hand, which slowed triage and made work-order quality inconsistent.

Trust in alerts

Unconfirmed anomalies risked being treated as confirmed failures, leading to unnecessary downtime or a loss of trust in alerts.

Intermittent connectivity

Vehicles connect over mobile networks, so the platform had to treat lost connectivity as normal and never mistake silence for a healthy or faulty vehicle.

Data protection and no KPI baseline

Vehicle, location and driver data needed encryption and role-based access, and time to detect, breakdown frequency and false-positive rate were not tracked.

No cost visibility

Messaging, streaming, inference, GenAI and storage costs could not be attributed to fleets, depots or stages, and no cost per vehicle existed.

The Solution

Rules, machine learning and GenAI working together, with people in charge

Indovision Services designed a serverless, streaming platform on AWS for Nuevas that follows six steps from vehicle telemetry to a reviewed work order.

  1. Telemetry ingestion

    Timestamped OBD-II / CAN bus data arrives through AWS IoT Core with per-device X.509 identities, then is buffered in Amazon Kinesis Data Streams and validated by AWS Lambda.

  2. Context and detection

    Records are normalized across makes and models and enriched with vehicle, trip and service history, while threshold rules and Amazon SageMaker AI anomaly scoring flag developing faults.

  3. Explanation

    Amazon Bedrock explains the observed signals, likely causes and data limitations in plain language, describing an anomaly as a potential fault rather than a confirmed failure.

  4. Work-order drafting

    A proposed work order is drafted with vehicle details, evidence, a suggested priority and approved manual references, with nearby-workshop context from Amazon Location Service.

  5. Alerting and review

    Amazon EventBridge and Amazon SNS route qualifying events to the responsible team, and staff approve before repairs are scheduled or a vehicle is restricted.

  6. Continuous improvement

    Confirmed repair outcomes and false-positive feedback tune rules, prompts and models under governance, with history kept in Amazon S3 and queried with Amazon Athena.

Fleet vehiclesOBD-II / CAN gatewaysMQTT over mutual TLSLocal buffering IngestionAWS IoT CoreKinesis Data StreamsAWS Lambda DetectionThreshold rulesSageMaker AI scoringHealth events GenAI explanationBedrock · work ordersKB · Guardrails Alerting & reviewAmazon EventBridgeAmazon SNS · SQSStaff approval DataAmazon S3Amazon DynamoDBAmazon AthenaLocation Service Apps & APIsAPI Gateway · ECS portal · Load Balancer Security & governanceIAM · Organizations · KMS · Config · CloudTrail ObservabilityAmazon CloudWatch · Amazon SNS Two Availability Zones · private subnets + VPC endpoints · pilot-light DR region · target RTO ≤ 4 h · RPO ≤ 15 min

Solution architecture: vehicle gateways stream telemetry into AWS IoT Core and Kinesis, rules and SageMaker AI detect faults, Amazon Bedrock explains them and drafts work orders, and staff review every alert before action.

People stay in control throughout. Explanations and draft work orders are recommendations, and the decisions to repair, schedule, restrict or withdraw a vehicle are never automated. Devices connect to AWS IoT Core over MQTT with TLS and mutual authentication, processing runs in private subnets across two Availability Zones, and VPC endpoints keep traffic to Amazon S3, DynamoDB, Kinesis, Bedrock and SageMaker AI off the public internet. AWS IAM, AWS Organizations, AWS Secrets Manager, AWS KMS, AWS Config and AWS CloudTrail govern the environment, with Amazon CloudWatch and Amazon SNS for observability.
The Outcome

Earlier warnings, planned maintenance and people in charge of every decision

With the platform Indovision Services designed, Nuevas Technologies expects its teams to detect developing vehicle faults earlier, plan maintenance before breakdowns occur and spend less effort preparing work orders, while keeping authorized staff in charge of every repair, scheduling and vehicle-restriction decision. Indovision Services delivered the foundation for these outcomes: telemetry streaming, rule and ML detection, evidence-grounded GenAI explanations, security designed in from the start, recovery planned around intermittent networks and cost controls that follow vehicles and events.

This case study describes an illustrative solution design and expected benefits only; no measured customer results are claimed. Targets are to be validated against baselines and measured results, and customer-specific details, thresholds, availability objectives and RTO/RPO values must be confirmed with Nuevas Technologies.

Day-to-day benefits

Earlier warning of developing faults

Rules and ML anomaly scoring surface potential faults from live telemetry, so maintenance can be planned before a breakdown.

Insights people can act on

Plain-language explanations state the signals, timestamps, likely causes and data limitations instead of raw diagnostic codes.

Less time on manuals and paperwork

Draft work orders arrive with evidence, a suggested priority and approved manual references for staff to review.

Better prioritization

Combined rule and ML evidence, enriched with trip, service-history and workshop context, focuses attention on the highest-risk vehicles.

Trustworthy alerts

Evidence-grounded generation, explicit uncertainty, Amazon Bedrock Guardrails and de-duplication keep anomalies distinct from confirmed failures.

People stay accountable

Authorized staff approve every repair, scheduling and vehicle-restriction decision. Nothing consequential is automated.

Secure by design and auditable

Per-device identities, least-privilege roles, KMS encryption, private networking and CloudTrail logging protect vehicle, location and driver data.

Alerts that keep working

Connectivity loss is expected and handled, and rule-based alerts continue even if ML or GenAI services are unavailable.

Room to scale, with costs in view

Serverless streaming services scale with the fleet, Bedrock runs only for qualifying events, and budgets and a per-vehicle baseline keep cost proportionate.

Want to plan fleet maintenance before breakdowns happen?

Indovision Services helped Nuevas move from reactive repairs to planned maintenance. We design and build secure generative AI solutions on AWS, from first architecture review to production.

Book a consultation