Use cases /

EAM · Asset Management

What Asset Management does, with agents.

Asset Management (EAM) keeps a single record for every physical asset an organisation runs on - laptops, monitors, network gear, furniture - from requisition through assignment, transfer, depreciation and disposal, with operational procurement wrapped around the same record. It is early and pre-release, and it carries no in-product AI of its own. The use cases below show how the Splenta platform agents are designed to work alongside it, illustratively rather than as delivered results.

A desktop computer and laptop with accessories on a bright office desk beside a window

Move routine requisitions through, escalate only the ones a person should see

Arin advances low-risk asset requisitions and surfaces the ones that need a human decision.

The situation

Requisitions pile up in a shared queue, each moving PENDING to APPROVED before it can become a purchase order. Most are routine repeat buys, but a few - a new vendor, an unusually large line, an off-catalog item - genuinely need someone to look.

How Arin helps

Arin is designed to watch the requisition queue and act within a scoped, audited mandate: advance requisitions that match known patterns and stay under agreed thresholds, and hold back the rest with the reason attached. A new vendor or a line above a limit becomes an exception in front of an approver rather than another row to triage. Every action Arin takes is logged against the requisition it touched.

What changes

The intent is that routine buying keeps moving without a person clearing each row by hand. Approvers spend their attention on the requisitions that actually carry a decision.

Printed financial documents laid out side by side on a desk for cross-checking, with a pen and coffee

Reconcile ordered, received and billed before an invoice is trusted

Arin compares requisition, purchase order, receipt and invoice line items and raises the mismatches.

The situation

The procurement flow records four documents for each buy - requisition, purchase order, receipt and purchase invoice - each with its own line items. A vendor bills for more than was received, or a receipt closes short of what was ordered, and the gap hides in the line detail.

How Arin helps

Arin is designed to read the line items across the four linked documents and check that what was ordered, received and billed line up. Where quantities or prices diverge - an invoice above the PO, a receipt that never completed - it surfaces the specific lines as an exception rather than silently passing them through. EAM records this paperwork but is not accounts payable or a general ledger, so Arin flags the discrepancy for a human and your ERP, and does not post anything.

What changes

Mismatches between the order, the goods and the bill are meant to surface before anyone acts on the invoice. The reconciliation stays as an operational check that sits alongside finance, not inside it.

A hand holding a calculator over financial statements and spreadsheets on a desk

Run the depreciation schedule and flag the book-value edge cases

Arin exercises the register's depreciation methods and surfaces assets that need a finance decision.

The situation

Asset groups set a depreciation method and useful-life defaults - straight line, declining balance or sum-of-years-digits - and any asset can override them. Across a large register, some assets approach salvage value, some carry an override that no longer matches their group, and those are easy to miss at period close.

How Arin helps

Arin is designed to walk the register on a schedule, compute book value using the method each asset actually carries, and compare the result against what the finance schedule expects. Assets nearing salvage, or with an override that has drifted from the group default, become a short exception list. Arin reports the arithmetic and the edge cases; it does not overwrite the numbers finance owns.

What changes

Book value is meant to stay in step with the schedule a finance team would run by hand. The judgement calls arrive as a reviewed list instead of being discovered after close.

Rows of server racks with neatly cabled IT hardware in an equipment room

Keep the register honest about idle, in-repair and end-of-life assets

Arin watches asset status and surfaces the ones due for transfer, repair follow-up or disposal.

The situation

Assets carry a status, a location, a holder and a useful life. Over time some sit IN STORE unused, some linger IN REPAIR longer than they should, and some pass the end of their useful life still on the active register.

How Arin helps

Arin is designed to read the register and act on the status and lifecycle fields already there. It can surface assets that have been IN STORE with no holder, ones stuck IN REPAIR past a reasonable window, and ones whose depreciation clock has run out but that were never disposed. Each becomes a suggested transfer or disposal for someone to confirm, attributed and audited, rather than an automatic change.

What changes

The register is meant to reflect what is actually in use, in repair or retired, without a periodic manual audit. Assets that should move, come back or leave the books get raised while they still matter.

A clean white desk with a fresh laptop, external monitor, keyboard and notebook set up for a new employee

Build a new-hire asset provisioning flow in plain language

PECO turns a described onboarding routine into a repeatable requisition-and-assignment workflow.

The situation

Every new hire needs a standard kit - a laptop, a monitor, the right peripherals - and today an IT lead re-creates the same requisition and assignment steps by hand each time, from memory.

How PECO helps

PECO is designed to take that routine in plain language - who gets what, which categories, which approver - and build it into a confirmed, repeatable workflow through its visual builder. The flow can raise the standard requisition, route it for approval, and set up the assignment against the new holder once the assets arrive. The person describing it reviews and confirms each step before it becomes an automation, so nothing runs that was not agreed.

What changes

A recurring onboarding routine is meant to become a workflow anyone can trigger rather than tribal knowledge. The same standard kit gets requested and assigned the same way every time.

A manager focused on a laptop while seated in a bright, modern office lounge

Let a trusted assistant read the live register and act within scope

Semorel lets an authorized AI agent understand the on-screen asset app and invoke only permitted actions.

The situation

A manager wants to ask an assistant plain questions about their own assets - what is assigned to a team, what is due for disposal - and occasionally kick off an action like raising a requisition or starting a transfer, without learning the app's screens.

How Semorel helps

Semorel is designed as a client-side SDK that lets a trusted AI agent read the live app's semantics - the assets, statuses and documents actually on screen - and invoke only the actions that user is authorized to take. The agent can answer from what the register really shows and, within scope, initiate a transfer or a requisition that still follows the app's normal approval path. It cannot reach past the permissions of the person it is acting for.

What changes

The register is meant to become answerable and actionable in plain language, bounded by each user's real authorization. Nothing the assistant does escapes the same scoping and approval the app already enforces.

See Asset Management run on your work.

These use cases are how Asset Management is designed to work with the Platform agents. Tell us about your operation and we will set up a look.