The challenge: managing assets at industrial scale
A large engineering conglomerate operating across multiple manufacturing facilities faced a problem common to organisations of its scale: no one actually knew where all the assets were. The asset register existed — thousands of rows in spreadsheets maintained by finance — but those spreadsheets were perpetually out of date. Equipment moved between departments without formal transfers. Maintenance history lived in paper logbooks, if it was recorded at all. Depreciation schedules ran on assets that had been scrapped years earlier.
Physical verification was the worst part. Conducting a full asset audit meant deploying teams of people across multiple facilities for several weeks. Auditors walked the shop floor with printed lists, manually matching asset tags to register entries, reconciling discrepancies by hand. By the time the audit was compiled and submitted, some of the data it contained was already stale. The organisation was in a continuous state of never quite knowing the true position of its fixed asset base.
The finance team had a specific concern: ghost assets. These are assets that appear on the books — still being depreciated — but which no longer physically exist. They had been scrapped, cannibalised, or simply lost, but the paperwork to remove them from the register had never been filed. For a large conglomerate with regulatory and audit obligations, this is not a minor inconvenience. It distorts depreciation calculations, inflates asset values, and creates exposure in statutory audits.
Maintenance was a separate but related failure. Without a reliable asset register, there was no foundation for planned preventive maintenance. Work orders were raised reactively — when something broke — rather than on a scheduled basis. The organisation had calibration and inspection schedules on paper, but no system to trigger reminders, track compliance, or link inspection records to specific assets.
The solution: tag everything, track the lifecycle
The implementation followed a clear principle: every physical asset gets a unique identity, and every event in that asset's life gets recorded against that identity. The execution required both hardware and software working in concert.
Assets were classified into two categories. Industrial assets — machinery, equipment, vehicles, material handling systems — received passive RFID tags capable of surviving the heat, vibration, and chemical exposure typical of manufacturing environments. Non-industrial assets — furniture, IT equipment, office fixtures — received standard barcodes. The distinction was purely practical: RFID tags cost more and are warranted where the reading environment is hostile; barcodes work fine where conditions are benign.
Field operators were equipped with handheld terminals running both Windows and Android. The terminals could read barcodes and RFID tags and sync with the central system over Wi-Fi. For areas without reliable connectivity, the terminals operated offline and queued transactions for sync when connection was restored. This was a non-negotiable requirement: manufacturing floors are not uniformly connected, and an asset tracking system that fails when Wi-Fi is patchy is useless in practice.
The system captured the complete asset lifecycle from the moment an asset entered the organisation:
- Procurement and tagging: New assets are tagged at goods receipt. Tag ID, asset details, and location are recorded before the asset leaves the stores area.
- Issue and custody: When an asset moves from stores to a department, the transfer is recorded against both the issuing and receiving cost centres.
- Transfers and movements: Any subsequent movement between locations, departments, or facilities requires a scan-based acknowledgement at both ends. No paper transfer forms.
- Maintenance and inspections: Every maintenance activity — whether scheduled PM or breakdown repair — is logged against the asset record with date, technician, and work description.
- Returns and disposals: When an asset is returned, scrapped, or sold, the transaction closes the asset record and removes it from active depreciation.
Preventive maintenance as a first-class concern
One of the explicit requirements from the engineering team was that asset tracking and maintenance scheduling should not be separate systems with a manual handoff between them. In the old setup, the ERP held asset records and a separate CMMS held maintenance schedules. Neither talked to the other reliably. Technicians checked one system for asset details and another for work orders, and the two were chronically out of sync.
In the implemented system, each asset record carries its own maintenance configuration: service intervals, calibration cycles, inspection checklists, and the identities of responsible technicians. The system generates work orders automatically when due dates approach, sends alerts to assigned technicians, and tracks completion. Critically, all maintenance records are stored as history against the asset — not as standalone work orders in a separate module.
This architecture makes it possible to query an asset's complete history in a single view: when it was acquired, every location it has been, every maintenance activity performed, every calibration certificate issued. For assets subject to statutory inspection requirements — pressure vessels, lifting equipment, electrical apparatus — this audit trail is not optional. It is a compliance obligation.
The system also tracks Mean Time Between Failures (MTBF) per asset class. Where MTBF data accumulates over time, it informs decisions about refurbishment versus replacement. A piece of equipment with an MTBF declining quarter-on-quarter is a candidate for replacement planning, not another round of reactive repairs.
The ghost asset problem resolved itself. Once the physical verification process was scan-based and continuous rather than annual and manual, assets that did not exist could not survive an audit cycle. Approximately 3–4% of the book asset value turned out to be ghost assets — equipment that had been disposed of but never formally written off.
Results: from weeks to minutes
The most visible improvement was in physical verification time. Previously, auditing a single large facility required teams of people over multiple days. With scan-based verification, a trained operator can cover a zone in under 20 minutes — scanning tags and letting the system automatically reconcile what was found against what should be there. Discrepancies are flagged in real time, not surfaced weeks later during report compilation.
The overall audit accuracy rate reached 97%, which is effectively the ceiling for a large multi-facility operation where a small number of assets are genuinely in transit or temporarily out of their assigned location at any given moment. The remaining 3% discrepancies are almost always explainable — assets legitimately checked out, equipment at an external service vendor — rather than genuinely missing or misrecorded.
For finance, the impact was immediate. The asset register is now a live document rather than a periodic snapshot. Depreciation runs on the actual asset population, not a stale spreadsheet. The ghost assets identified during the initial tagging exercise were written off, correcting years of accumulated distortion in the books. Subsequent audits by statutory auditors have proceeded without the multi-week preparatory exercise that used to precede them.
For operations, the maintenance compliance rate improved substantially. When work orders are generated automatically and technicians receive direct alerts, scheduled maintenance no longer gets displaced by the urgency of reactive repairs. The organisation has moved from a primarily reactive maintenance posture to one where the majority of equipment interactions are planned. The financial value of that shift — in reduced downtime, extended asset life, and lower emergency repair costs — is difficult to isolate precisely, but the direction is unambiguous.