
NefteStroyService Ltd. is one of the largest construction and service companies in Western Kazakhstan. It provides a full service cycle: from construction and capital repairs to maintenance of oil and gas facilities.
01
Client context
- A large fleet of equipment across various types and categories
- A broad geography of projects and bases
- A complex structure of divisions and contractors
- A high share of manual processes and Excel spreadsheets
02
Operational challenges
- Difficult to monitor equipment condition, location and workload
- Maintenance schedule violations led to downtime and breakdowns
- No unified record of spare parts and materials across sites
- No unified history of requests, repairs, approvals and downtime
03
What was implemented
- Integration with GPS and fuel-level sensor (ДУТ) devices
- Detailed registration of the entire equipment fleet
- Maintenance schedule based on mileage and regulation control
- Defect log and work-order module
- Repair provisioning (inventory requests, warehouse)
- A mobile app for drivers and technical staff
- Manager and dispatcher dashboards, KPI analytics
- Integration with 1C: end-to-end expense tracking
04
What became manageable
- A unified digital loop for fleet management
- Full transparency of repairs, downtime and costs
- Maintenance discipline and reduced breakdown rate
- Control over equipment workload and allocation across projects
- End-to-end analysis of utilization and project cost
- A digital and transparent inventory-provisioning process
Results achieved
−82
Units of equipment — based on redundancy analysis (this is only the beginning)
−14%
Fuel consumption — thanks to automatic overspend alerts and analytics
−23%
Reduced downtime — through even workload distribution and system-based requests
8 → 2
Reduction of average repair downtime from 8 to 2 days
15%
Improved maintenance discipline; scheduled-maintenance failures reduced through mileage-based forecasting and alerts
85%
False fuel-theft alerts reduced through fuel-sensor calibration and improved algorithms