The softwarelayer overyour hardware
28 modules across six industries, on one core. Device orchestration, serialisation, rules and analytics — so RFID stops being a project and starts being a system.

A project ends.
A platform compounds.
Most RFID in this market is delivered as a one-off. A reader here, a tag estate there, a spreadsheet in the middle, and a supplier who has moved on by the time it needs changing.
That model works exactly once. The second use case cannot reuse the first one’s data, the integration is rebuilt from scratch, and nobody can answer a question that spans two systems — because there is no shared model of what a tag, an item or an event actually is.
A platform inverts the cost curve. The first module carries the device layer, the tag registry, the rules engine and the connectors; every module after it inherits all four. That is why a fitting-room deployment on top of existing inventory tagging is a fraction of the cost of the inventory deployment itself, and why the return-fraud case is even possible — it needs the POS link that inventory already built.
- 01
Tags
What carries the identity
- UHF RAIN — retail, WIP, logistics, library, linen
- On-metal — tools, moulds, equipment
- Laundry — linens and textiles
- Tamper-evident — returns and pharma
- HF/NFC — access, patient ID, unit-dose
- Passive sensor tags — temperature and shock
- 02
Readers & edge
What turns presence into an event
- Fixed portals and gates
- Overhead arrays
- Desktop encoders
- Handheld readers
- Smart cabinets
- Edge gateways with 4G/5G backup
- 03
Platform
What decides what the event means
- Device orchestration and health monitoring
- Tag registry and serialisation
- Event ingestion, normalisation and deduplication
- Rules engine and workflow automation
- Role-based access and audit trails
- Dashboards, alerting and KPI tracking
- 04
Integrations
What acts on the decision
- REST and GraphQL APIs
- Prebuilt connectors
- Webhooks
- Offline store-and-forward
- BI and data export
What every module
inherits
Six services you build once. Nobody buys these on their own — they are why the second module is cheap.
- 01
Device orchestration
Onboarding, firmware, remote configuration and health monitoring for every reader, portal, encoder and gateway on site — so a dead antenna surfaces as an alert rather than as a gap in the data three weeks later.
- 02
Tag registry & serialisation
One identity per item, linked to your item master, with an EPC memory policy that stays consistent across sites. This is the layer that makes return validation and chain-of-custody possible at all.
- 03
Event processing
Millions of reads a day per site, normalised and deduplicated into events that mean something. A tag read four hundred times as someone walks past is one event, not four hundred.
- 04
Rules engine
Where business logic lives — return eligibility, access policy, reorder thresholds, PM triggers, exception routing. Configured per customer rather than coded, so a policy change is not a release.
- 05
Integration layer
Prebuilt connectors plus documented APIs, with offline buffering and store-and-forward so a dropped link degrades gracefully instead of losing the day's reads.
- 06
Analytics & reporting
Role-based dashboards, scheduled reports and SLA tracking, with export to whatever BI tool you already run. Operations, finance and compliance each get their own view.
Six industries
Same core underneath. What changes is the tag, the read point and the system it has to talk to.
- 8 modules
Retail & E-commerce
Shrink, return abuse and stock that the system says is there. Eight modules on one platform.
98–99%Inventory accuracy - 5 modules
Manufacturing
WIP you cannot see, schedules that slip, and maintenance that happens after the breakdown.
15–25% lowerWIP dwell time - 5 modules
Healthcare & Pharma
Medication safety, patient identification and counterfeit control — deployed ward by ward.
50–85% lowerMedication errors - 5 modules
Education
Annual audits that take weeks, assets that quietly walk, and a library still stamping cards.
70–90% lowerAudit time - 2 modules
Logistics
Pallets, containers and yard moves — where loss is measured in whole units.
Materially lowerUnit loss - 3 modules
Hospitality
Linen that disappears, amenities used by people who did not pay, assets nobody logs.
30–50% lowerLinen loss
It has to talk to
what you run
Connector delays, not RF, are what actually holds these projects up. Prebuilt beats bespoke every time.
ERP
SAP, Oracle, Microsoft and local platforms
POS
Transaction, deactivation and return validation
WMS / TMS
Putaway, picking and transport events
MES
Station routing, BOM and WIP status
LMS / ILS
Library circulation and campus systems
HIS / eMAR
Clinical records and medication administration
PMS / BMS
Hospitality property and building systems
CMMS / TOS
Maintenance management and terminal operations
Where a prebuilt connector does not exist, integration runs through documented REST and GraphQL APIs, or a scheduled file exchange for systems that offer nothing else.
What it moves
Ranges across deployed sectors, not a single best case. Your own numbers are measured against a baseline captured before install.
- Inventory accuracy
- 98–99%From 65–80% typical
- Shrink reduction
- 20–40%
- Audit time
- 70–90% less
- Unplanned downtime
- 15–30% less
A pilot, then
a decision
Ninety to a hundred and twenty days, with the scale decision gated on measured KPIs rather than on how the demo went.
- 01
Survey & baseline
Weeks 0–2
RF site survey, read-zone design, success criteria agreed in writing, and four to eight weeks of baseline KPI capture started.
- 02
Install & configure
Weeks 3–6
Hardware installed, platform configured, connectors built and tested, UAT signed off.
- 03
Go-live & hypercare
Weeks 7–10
Controlled go-live with elevated support, user training, SOP changes embedded on the floor.
- 04
Verify & decide
Weeks 11–12
KPIs measured against baseline, ROI summary issued, scale decision made on evidence.
You leave a pilot with an RF design pack, an integration checklist, training materials, a KPI report against baseline, and a scale plan with a bill of materials — whether or not you decide to scale.
What you can hold us to
| Tier | Hours | Uptime | Response |
|---|---|---|---|
| Silver | 8×5 | 99.5% | P1 in 2 hrsP2 in 4 hrs, P3 next business day |
| Gold | 12×6 | 99.9% | P1 in 1 hrProactive monitoring, quarterly health checks |
| Platinum | 24×7 | 99.95% | P1 in 30 minsDedicated CSM, DR tests, annual roadmap workshop |
Designed in, not added
- Data minimisation
- Tags carry identifiers, never personal data. Anything sensitive lives server-side behind role-based access.
- Encryption
- In transit and at rest, with key rotation and encrypted tag memory where it applies.
- Audit trails
- Immutable logs of who saw what and who changed what — built for clinical governance review, not just for IT.
- Data residency
- Configurable region, with on-premise edge buffering for sites that will not accept cloud-only.
Before you
commit
Do we have to buy the whole platform?
No. You buy modules. They share one core — device management, tag registry, rules, analytics — which is why the second module costs materially less to deploy than the first, and why they can talk to each other rather than sitting in silos.
We already have RFID hardware. Does this replace it?
It sits above it. The platform is a software layer over readers and tags, including ones we did not supply. Where existing equipment cannot be orchestrated remotely we will say so during the survey rather than after the invoice.
What happens when the internet drops?
Edge gateways buffer locally and forward when the link returns, and critical flows — EAS deactivation, door access — have local rule fallbacks so they keep working offline. This is designed for Bangladeshi connectivity, not assumed away.
How long before we see a number?
A pilot runs 90 to 120 days, and the KPI readout lands in weeks 11–12 measured against a baseline captured before anything was installed. If the delta is not there, the scale decision is a no — that is the point of gating it.
Is our data used to train anything?
Not without a written agreement. Where benchmarking across customers is offered it is anonymised and opt-in, and it is never the default.
Start with onemodule, one site,one number
- Call+880 1713 819528 · 01713 222558
- Emailservice@ratsbd.com
- VisitGround Floor, House-7, Road-13, Sector 14, Uttara, Dhaka-1230
Get your quote
Tell us your store type and city. A specialist surveys your floor, recommends the exact system and sends a clear quote — no obligation.
Request a quote Chat on WhatsApp