Home / Solutions / Multi-Site Unified Monitoring Platform

Multi-Site Unified Monitoring Platform

Multi-site unified monitoring should first establish unique codes for sites, gateways, devices and points, then unify the upload protocol, time basis and alarm grading. CW or ESX handles field access, and FEXLINK handles centralized monitoring, permissions and reports. Whether the platform supports a given site scale, deployment mode and third-party interfaces must be confirmed against the actual configuration; project capability cannot be estimated from site count alone.

Device-Edge-Cloud CollaborationCentralized Monitoring

Solution overview

The solution is for parks, distributed energy, water or outdoor facilities managed across many sites. The focus is to establish a scalable site tree and point semantics once, instead of each site using its own naming and alarm rules.

Applicable industries

  • Multi-industry industrial sites
  • Industrial parks
  • Water and wastewater treatment

Primary user roles

  • Automation engineers
  • System integrators
  • O&M staff

Problems addressed

  • Sites are distributed and alarms and O&M status are not unified
  • Device naming, point units and alarm levels differ between sites
  • Headquarters, regional and site staff need different scopes of viewing and response permission

Recommended product combination

ESX

ESX Smart Edge Computing Gateway

Handles device access, edge computing and upload at some sites; device scale, points and network capability are confirmed by specific model

FEXLINK

FEXLINK IoT Platform

Provides the site tree, device models, alarm center, permissions, history and reports

System architecture and data flow

Multi-Site Unified Monitoring Platform System TopologyFour-layer data flow across perception, edge control, platform and application for Multi-Site Unified Monitoring Platform.01Perception layer
Field instruments, controllers and subsystems at each site connect to an edge gateway or ESX through project…
02Edge / control layer
Edge devices perform acquisition, point mapping, necessary buffering and local alarms, and backfill after ne…
03Platform layer
FEXLINK organizes data by organization, region, site, device and point, and configures unified alarms and pe…
04Application layer
Provides multi-site overview, site details, alarm response, historical queries and reports, with each role v…

Perception layer

Field instruments, controllers and subsystems at each site connect to an edge gateway or ESX through project-confirmed interfaces

Edge / control layer

Edge devices perform acquisition, point mapping, necessary buffering and local alarms, and backfill after network recovery

Platform layer

FEXLINK organizes data by organization, region, site, device and point, and configures unified alarms and permissions

Application layer

Provides multi-site overview, site details, alarm response, historical queries and reports, with each role viewing its own scope

Data flow

  1. Establish unique codes for organization, region, site, gateway, device and point
  2. Edge devices acquire site data and upload it to the platform, retaining necessary buffers during anomalies
  3. The platform evaluates online status, limit violations, communication and combined events under unified rules
  4. Alarms are assigned by site and responsible person, recording acknowledgement, handling and closure status
  5. The platform summarizes cross-site availability, alarm trends and data quality

Integration and interfaces

  • Field devices connect through project-confirmed Modbus, RS485, Ethernet or DI/DO interfaces
  • Edge-to-platform links can use project-confirmed MQTT or HTTP API, with device identity and access control configured
  • Existing enterprise systems can share site and alarm data through API; data direction and permissions must be confirmed
  • Private deployment requires defining servers, databases, backup, network and O&M responsibility

Applicable standards

  • Applicable project standards for industrial monitoring, data communication and cybersecurity; the final edition is confirmed against project location, industry and project documents
  • Field communication uses project-confirmed protocols such as Modbus; platform interfaces use MQTT or HTTP API

Acceptance and delivery boundaries

Suggested acceptance dimensions

  • Check organization, site, gateway, device and point codes to ensure no duplicate objects across sites
  • Verify the visibility scope and operation permissions of headquarters, regional and site accounts by role
  • Simulate site offline, gateway network loss, alarm recovery and backfill, and check platform status and records
  • Sample-check data consistency in cross-site overview, site details and historical reports

Limitations and exclusions

  • The platform itself cannot repair site networks, power supplies or field device faults
  • Site totals, concurrency, history retention and interface quotas depend on actual deployment and version; page examples cannot be used to estimate them
  • Data sharing between organizations requires explicit authorization and is not visible across tenants by default
  • Unified monitoring is not unified control; remote operation requires separate permission and safety interlock design

Operational value

  • Brings distributed sites into a unified device, alarm and O&M view
  • Reduces duplicate configuration and unauthorized access through standard codes and a permission model
  • Gives cross-site data quality and anomaly response a unified record

Selection guidance

  • Define the number of sites, device scale per site, upload period, data residency and account scope before choosing SaaS or private deployment
  • Where networks are unstable, configure buffering and backfill first; local alarm requirements should be pushed down to the edge layer
  • The multi-site hierarchy should allow for region and tenant dimensions, but the organization tree should not be extended without a business need
  • For map or topology display, confirm coordinates, base map, permissions and update responsibility separately

Typical implementation patterns

  • Typical implementation pattern: a park or distributed facility builds a site tree by region, each site uploads devices and alarms through a gateway, the platform shows the visible scope by O&M role, and acceptance covers network-loss and cross-site query scenarios. This is an anonymized typical pattern and does not represent specific customer site scale.

Before retrofit → With the solution

Before retrofit

  • Sites are scattered; alarms and O&M status are not unified; cross-site comparison relies on manual aggregation
  • Different sites have inconsistent device naming, point units and alarm levels, making unified platform display difficult
  • Headquarters, regional and field personnel need different scopes of viewing and handling permissions; existing systems do not support tiered access

With the solution

  • Establishing unique IDs for organization, region, site, gateway, device and point
  • CW or ESX handles field access, edge computing and upload
  • FEXLINK organizes device models, alarm center, permissions and history by site tree

How to evaluate the effect

01

Measurable signal

Site Online Rate

02

Data source

The platform calculates the ratio of sites with normal communication to expected-online sites

03

Review method

Online rate change statistics by region and time window

04

Acceptance boundary

The platform itself cannot fix site network, power supply or field device faults

At site-ID granularity; sites with normal communication heartbeat counted as online

Frequently asked questions

Must a multi-site platform use SaaS?

No. Cloud service or local deployment can be chosen depending on data residency, network, O&M and account requirements. The interfaces and responsibility boundaries differ.

Can headquarters see every site?

Visibility is determined by role and data permission. The platform should support authorization by organization, region or site and retain viewing and operation audit records.

Can the platform still show data after a site loses network?

During an outage the platform usually cannot receive data in real time; whether data is backfilled after recovery depends on edge buffering. Local site alarms should be handled at the site wherever possible.

How are alarms unified across sites?

Device types, point semantics, alarm levels and naming rules must be unified first and then mapped on the platform. Reusing each site's original alarm names hinders cross-site comparison.

Can it connect to the customer's existing platform?

Yes, through project-provided API or messaging interfaces, after confirming the data dictionary, authentication, frequency, network and both parties' responsibilities.

Request a matched product combination and project review

Share the application, circuit conditions and existing system interfaces so engineers can confirm selection and implementation boundaries.

Contact FEXLINK