What it does
Enough to run the work, and an honest line around what it is not.
AMS covers inspections, defects, risk and the registers those depend on. It does not try to be your planning and costing system, because you already have one.
Risk management
Risk scoring your engineers can change without us.
This is the deepest module in AMS, and the one configured most heavily at the process plant deployment.
Variables defined per site
A risk variable is a named input with its own evaluation expression. The expression is real arithmetic, not a dropdown of preset bands, so a site that scores consequence differently from the site next door simply configures it differently. Changing the model is configuration, not a release.
Measurements logged over time
Measurements are recorded against equipment and locations and kept as a log, so a reading is a point in a history rather than a value that overwrites the last one. The log has a bulk edit for when a campaign lands a hundred readings at once, and an overview for reading the estate rather than the row.
Assessments across the estate
Assessments roll the variables and measurements up against the equipment and location hierarchy, so the question “what is the risk under this area” has an answer that follows the structure you already maintain.
The grid above is the shape of a risk model, not a screenshot. In AMS the score is whatever your configured formula evaluates to.
Inspection authoring
Build the round in the product, not in our source control.
The Builder is where a reliability engineer authors what the field crew will be asked, and what should happen when they answer.
Questionnaires and questions
Author a round, order its questions, and set the limits each one is judged against.
A shared question catalogue
Questions are drawn from a catalogue rather than retyped per round, so the same check means the same thing everywhere it appears.
Per-site variants
One master round, adapted per site. Two plants running the same equipment can diverge where they need to without forking the round.
Question actions
The mechanism that turns a finding into work: when an answer meets its condition, AMS raises a defect notification already carrying the object part, damage code, priority and standing comments.
Mass edit
Change a limit or a code across many questions at once, rather than opening them one at a time.
One-pager export
The authored round as a single sheet, exportable to Excel for review and sign-off away from the screen.
Defects and work
From a finding to a scheduled operation.
Notifications
The defect record. Raised from an inspection answer by a question action, or raised directly by anyone who sees something — on the web, or from a phone in the plant. Attachments and annotated photos travel with it.
Work orders
A register with detail and search, so the work that came out of inspection findings is visible next to the rest of it.
Operations on a time axis
The operations scheduler places work order operations on a shared timeline across the work centres you select, with a second view that filters by order type. It is a planning board for seeing load, not a replacement for your scheduler.
Bulk import
Works Order Import maps spreadsheet columns onto work order, operation and task fields, validates against master data, previews what will be created, and reports the result.
AMS does not replace your preventive maintenance scheduler and does not try to own your planning and costing. There is no PM plan authoring screen in AMS, and we are not going to pretend otherwise in a demo.
What AMS does is feed that scheduler: it produces properly coded defect notifications, with evidence attached, from rounds that were actually executed. Where a site has asked for it, those notifications have been pushed into SAP Plant Maintenance as delivered integration work.
The registers underneath
Equipment and organisation, structured the way the plant is.
Inspections and risk have to hang off something. AMS keeps its own register of the assets you inspect and the structure they sit in — held through the DDF API, not in a database of ours.
Equipment and equipment types
The asset register, with a type classification that lets a question catalogue and a risk model apply to a class of equipment rather than one tag at a time.
Company, area and location
The hierarchy the rest of the product reports against. Assessments and searches follow it, so “everything under this area” is a real query rather than a manual list.
Employees and org positions
Who works where, and which position an inspection round is assigned to. Assignment is by position rather than by person, so a roster change does not orphan the work.
Reporting and finding things
Dashboards you already know how to build.
Embedded Power BI
AMS embeds Power BI reports directly in the application, authenticating and minting embed tokens server-side. Your analysts build reports in the tool they already use and your people open them without leaving AMS.
Search
One box across notifications, equipment, locations and work orders, with results grouped by type. It is the first item in the menu on every site.
Timeline
Work order operations placed on a shared time axis, for reading sequence and overlap rather than totals.
Excel export
Registers and logs export from the grid — the employee register, org positions, work order search results, the risk measurement log, the scheduler and the Builder one-pager.
Attachments and annotation
Every record can carry files and photographs. The built-in image editor lets someone mark up a photo — circle the crack, point at the leak — before it is attached, on the desktop and on the phone.
Access
Roles mapped to functions, not to screens.
AMS separates users, system roles and system functions, and maps roles to functions explicitly. A role is a set of things a person may do; screens follow from that rather than being the unit of permission.
Navigation itself is data-driven per site, so two customers on the same build see different menus. Sign-in is Microsoft Entra ID where you use it, and username and password where you do not.
White-labelling is structural rather than bolted on: the menu, the module set and the branding are configuration. That is how one product serves a smelting group and a process plant without either of them running a fork.
Ask us to show it against your equipment
Bring an inspection round you actually run and an asset hierarchy you actually have. That is a more useful hour than a scripted walkthrough.