Predictive Maintenance (Failure Prediction, RUL)

Supervised ML/DL on run-to-failure data and survival models: Aspen Mtell (claims failures detected up to 90 days ahead), SAP Failure Curve Analysis, Siemens Senseye, C3 AI Reliability.

time series

1named models

67candidates

13deployments

4in Indonesia

Named Models

System Invariant Analytics Technology (SIAT)NECReleasednot in the notesRuns onnot in the notesClasstime seriesNEC predictive maintenance and real-time equipment monitoring technology; proof of concept at Kalbe Farma (Apr to Jun 2026) met 2 of 3 targets per Kemenperin.Commercial use: proprietary terms; Jakarta hosting: unknown; Indonesian support: not stated

Where this comes from

Models the research names for this use case. Research, with sources on each model record.

Who Runs It

production, 10contract signed, 1proof of concept, 1planned, 1

PT Kilang Pertamina InternasionalCountry: IndonesiaPPMS Predictive and Prescriptive Maintenance System: machine learning anomaly detection on compressors and pumpsVendor: not statedproductionDate17 Oct 2025Described as existing; sites and results not statedEvidence: company statementConfidence: lowSource: investortrust.id

PT PLN (Persero) with IBMCountry: IndonesiaIBM predictive maintenance for generation and distribution assets (product not named); AI for smart grid load balancing under discussionVendor: IBMproductionDate9 Mar 2024In use for about 10 years according to IBM; smart grid AI plannedEvidence: company statementConfidence: lowSource: fortuneidn.com, fortuneidn.com

PETRONASCountry: MalaysiaAVEVA Predictive Analytics on Azure (about 200 models)Vendor: AVEVA; integrator Trisystem EngineeringproductionDate26 Oct 202151 major early warnings and RM73.1 million value in 2020; expanded to 10 more plants and 150 equipment trains (vendor-reported)Evidence: company statementConfidence: lowSource: frontier-enterprise.com

OCP EcuadorCountry: EcuadorAspen MtellVendor: AspenTech (Emerson)productionDate–Evidence: company statementConfidence: lowSource: aspentech.com

YPFCountry: ArgentinaAspen MtellVendor: AspenTech (Emerson)productionDate–Evidence: company statementConfidence: lowSource: aspentech.com

DowCountry: not in the notesC3 AI ReliabilityVendor: C3 AIproductionDate–Vendor claim: $45M per yearEvidence: company statementConfidence: lowSource: c3.ai

Flint Hills ResourcesCountry: USAC3 AI ReliabilityVendor: C3 AIproductionDate–Evidence: company statementConfidence: lowSource: c3.ai, en.wikipedia.org

HolcimCountry: not in the notesC3 AI ReliabilityVendor: C3 AIproductionDate–Vendor claim: 80 plants and 400 assetsEvidence: company statementConfidence: lowSource: c3.ai

ShellCountry: globalC3 AI ReliabilityVendor: C3 AIproductionDate–Vendor claim: Over 10,000 assetsEvidence: company statementConfidence: lowSource: c3.ai

TalgoCountry: SpainTSMART train predictive maintenance (more than 30,000 signals a second) and the Sophia maintenance agent on Vertex AIVendor: Google CloudproductionDate–No figures statedEvidence: company statementConfidence: lowSource: cloud.google.com

TASNEECountry: Saudi ArabiaGE Vernova SmartSignalVendor: GE Vernovacontract signedDateMar 2024Evidence: company statementConfidence: lowSource: gevernova.com

PT Kalbe Farma TbkCountry: IndonesiaSystem Invariant Analytics Technology (SIAT)Vendor: PT NEC Indonesia with Bogor Technology; NECproof of conceptDate20 Jul 2026Proof of concept Apr to Jun 2026, facilitated by Kemenperin PIDI 4.0; 2 of 3 targets met; Kalbe satisfaction 8/10Evidence: government statementConfidence: mediumSource: suarapemerintah.id

PT Pertamina International Shipping with Pertamina Marine EngineeringCountry: IndonesiaSIMONTANA mooring and port facility predictive maintenance, being upgraded with AI and IoT; AI real-time vessel monitoring already in useVendor: not statedplannedDate9 Dec 2024Vessel monitoring in use; the SIMONTANA AI upgrade is in progressEvidence: company statementConfidence: lowSource: jpnn.com

Where this comes from

Documented uses by named organisations. The evidence column says who made the claim.

Candidate Models

TSB-AD

A general benchmark for this model class, not a test on this use case. Time-series anomaly detection, 1,070 series from 40 datasets (VUS-PR); NeurIPS 2024; open to community submissions from 1 Apr 2026. Higher is better.

PaAnoSungkyunkwan University vendor-reported0.53 (Sub)-PCA independent0.42

MOMENT (MOMENT-1-large)Carnegie Mellon University independent0.39 DeepAnT independent0.31 OmniAnomaly maintainer-run0.31 PCA independent0.31

00.51

TimesFM 3.0Google, time seriesReleasedAug 2026open weightshas scoreslast 18 months

Toto 2.0 1BDatadog, time seriesReleasedJun 2026open weightshas scoreslast 18 months

Toto 2.0 2.5BDatadog, time seriesReleasedJun 2026open weightshas scoreslast 18 months

Toto 2.0 313MDatadog, time seriesReleasedJun 2026open weightshas scoreslast 18 months

Granite PatchTST-FM r1IBM, Rensselaer Polytechnic Institute, time seriesReleasedMar 2026open weightshas scoreslast 18 months

Moirai 2.0 smallSalesforce, time seriesReleasedNov 2025open weightshas scoreslast 18 months

Chronos-2Amazon, time seriesReleasedOct 2025open weightshas scoreslast 18 months

FlowState r1.1IBM, time seriesReleasedSep 2025open weightshas scoreslast 18 months

TimesFM 2.5Google, time seriesReleasedSep 2025open weightshas scoreslast 18 months

TiRexNX-AI, time seriesReleasedMay 2025open weightshas scoreslast 18 months

Showing 10 of 67. All models of this class

Where this comes from

Derived, not researched: models in the directory whose class matches a class this use case uses, excluding family records and models shut down or deprecated. Ranked by how many fit facts they meet, then by release date. Each fit fact is read from the model record.

Use Case

Industry: industrial and asset-intensive
Model classes: time series
Benchmarks or datasets: C-MAPSS turbofan degradation PRONOSTIA / FEMTO (IEEE PHM 2012)PHM Society Data Challenges
Regulatory notes: Surat Edaran Menkominfo 9/2023 Etika Kecerdasan Artifisial, Kementerian Komunikasi dan Informatika (now Kementerian Komunikasi dan Digital)Predictive maintenance (failure prediction, RUL) is an AI-based service run by an electronic system operator, and the circular asks such operators to build and use AI on the values of inclusivity, transparency, humanity and security.

Where this comes from

From the industry tracks T08 to T10 of the research notes.

R&D Topics

Links to R&D topics are awaiting review.

Where this comes from

Which R&D topics would most change this use case. Drafted by Claude from the topic definitions and the use case; proposed until approved.

In Neighbouring Countries

Tenaga Nasional Berhad (TNB)Malaysia, productionIn-house AI predictive maintenance for the 11kV underground cable network (ASEAN award)

Where this comes from

Documented deployments of this use case in Malaysia, Singapore, Thailand, Vietnam and the Philippines, each with its source.

Opportunities

Opportunities are awaiting review.

Where this comes from

Gaps found by rule in what the directory holds for this use case. Each is a proposal for review.

Links

IndustryIndustrial30 use cases, 54 deployments

Use CasePredictive Maintenance (Failure Prediction, RUL)Industrial

Models

System Invariant Analytics Technology (SIAT)NEC, time series

Who Runs It

PT Kalbe Farma TbkProof of concept, 20 Jul 2026

PT Kilang Pertamina InternasionalIn production, 17 Oct 2025 PT Pertamina International Shipping with Pertamina Marine EngineeringPlanned, 9 Dec 2024

PT PLN (Persero) with IBMIn production, 9 Mar 2024+9 more

Graph

Models 1

System Invariant Analytics Technology (SIAT)NEC, time series

Deployments 13

PT Kalbe Farma TbkSystem Invariant Analytics Technology (SIAT), Indonesia DowC3 AI Reliability Flint Hills ResourcesC3 AI Reliability, USA HolcimC3 AI Reliability OCP EcuadorAspen Mtell, Ecuador PETRONASAVEVA Predictive Analytics on Azure (about 200 models), Malaysia PT Kilang Pertamina InternasionalPPMS Predictive and Prescriptive Maintenance System: machine learning anomaly detection on compressors and pumps, Indonesia PT Pertamina International Shipping with Pertamina Marine EngineeringSIMONTANA mooring and port facility predictive maintenance, being upgraded with AI and IoT; AI real-time vessel monitoring already in use, Indonesia PT PLN (Persero) with IBMIBM predictive maintenance for generation and distribution assets (product not named); AI for smart grid load balancing under discussion, Indonesia ShellC3 AI Reliability, global TalgoTSMART train predictive maintenance (more than 30,000 signals a second) and the Sophia maintenance agent on Vertex AI, Spain TASNEEGE Vernova SmartSignal, Saudi Arabia

YPFAspen Mtell, Argentina

Sources

Aspen Mtell | AI-powered Prescriptive and Predictive Maintenance Solution | AspenTech aspentech.com https://www.aspentech.com/en/products/apm/aspen-mtell
Siemens Expands Industrial Copilot with AI-Powered Maintenance Solution etmm-online.com https://www.etmm-online.com/siemens-industrial-copilot-ai-powered-maintenance-solution-a-76790d2aebb90810d857abf89c94c051/
Cut Unplanned Downtime by up to 50% with AI c3.ai https://c3.ai/products/c3-ai-reliability/
arxiv.org/pdf/2511.19124 https://arxiv.org/pdf/2511.19124
Industri Farmasi Masuk Era AI, Kemenperin Fasilitasi Uji Coba Solusi Industri 4.0 suarapemerintah.id https://suarapemerintah.id/2026/07/industri-farmasi-masuk-era-ai-kemenperin-fasilitasi-uji-coba-solusi-industri-4-0/

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Predictive Maintenance (Failure Prediction, RUL)", https://aurion.semai.id/usecases/ind-predictive-maintenance/, records of October 2026.