Graha Surveyor Indonesia, Setiabudi, Jakarta Selatan corporate@liceoglobal.com +62 21 2966 6700
Senin–Jumat: 08.30 – 17.30 WIB Monday–Friday: 08.30 – 17.30 WIB
Manufaktur · Industry 4.0Manufacturing · Industry 4.0

Sensorisasi dan predictive maintenance untuk produsen otomotif tier-1Sensorization and predictive maintenance for a tier-1 automotive supplier

SektorSector
Komponen Otomotif · 3 pabrik Jawa BaratAuto Components · 3 West Java plants
DurasiDuration
12 bulan · 202412 months · 2024
Tim LiceoLiceo Team
8 profesionalprofessionals
Nilai EngagementEngagement Value
USD 2,1 M

KonteksContext

Produsen komponen otomotif tier-1 dengan tiga pabrik di Jawa Barat menghadapi unplanned downtime yang merugikan, terutama pada mesin stamping dan welding kritis. Setiap jam downtime senilai ~USD 8.000.A tier-1 automotive component manufacturer with three West Java plants faced costly unplanned downtime, especially on critical stamping and welding machines. Each downtime hour cost ~USD 8,000.

Tantangan UtamaPrimary Challenges

  • Unplanned downtime tinggi pada mesin kritisHigh unplanned downtime on critical machines
  • Maintenance crew terbatas; pemeliharaan reaktifLimited maintenance crew; reactive maintenance
  • Data sensor belum dimanfaatkan untuk prediksiSensor data not yet utilized for prediction
  • Skeptisisme tim setelah dua proyek serupa gagalTeam skepticism after two similar projects failed

PendekatanApproach

Pemetaan Biaya Downtime (Bulan 1)Downtime Cost Mapping (Month 1)

Hitung biaya downtime per mesin per jam dalam kolaborasi finance + maintenance. Identifikasi 84 mesin kritis (threshold >USD 5.000/jam).Calculate downtime cost per machine per hour in finance + maintenance collaboration. Identify 84 critical machines (threshold >USD 5,000/hour).

Instalasi Sensor (Bulan 2–3)Sensor Installation (Months 2–3)

Instalasi sensor vibrasi, akustik, dan suhu pada 84 mesin. Validasi data stream ke cloud lakehouse.Vibration, acoustic, and temperature sensor installation on 84 machines. Validate data streams to cloud lakehouse.

Baseline & Model (Bulan 4–9)Baseline & Model (Months 4–9)

Pengumpulan baseline data selama 6 bulan. Pengembangan model ML production-ready, dilatih ulang setiap minggu.6-month baseline data collection. Production-ready ML model development, retrained weekly.

Integrasi Workflow (Bulan 10–12)Workflow Integration (Months 10–12)

Integrasi alert ke work order system & mobile app teknisi. Pelatihan 24 teknisi. Dashboard supervisor & eksekutif.Alert integration to work order system & technician mobile app. Training for 24 technicians. Supervisor & executive dashboards.

Hasil TerukurMeasurable Outcomes

Unplanned downtime turun 64% dalam 12 bulanUnplanned downtime reduced 64% in 12 months
Biaya pemeliharaan turun 31% — tanpa mengurangi reliabilitasMaintenance cost reduced 31% — without sacrificing reliability
Payback investasi tercapai dalam 9 bulanInvestment payback in 9 months
Praktik direplikasi di 4 pabrik lain dalam grupPractice replicated across 4 sister plants
Tim maintenance pindah dari reaktif ke strukturalMaintenance team moved from reactive to structural work

Diskusi pendekatan serupa untuk organisasi AndaDiscuss similar approach for your organization

Detail engagement dapat dibagikan lebih lanjut setelah kesepakatan kerahasiaan.Engagement details can be shared further after a confidentiality agreement.

Hubungi KamiContact Us