MRM Compliance Dashboard — Power BI Build Guide

Overview

Examiner-facing dashboard for model inventory status, validation coverage, risk distribution, and monitoring trends. This dashboard supports compliance with Fed SR 11-7 and OCC 2011-12 reporting expectations by providing centralized visibility into MRM program metrics.

Note: This dashboard aids in meeting regulatory reporting requirements. Organizations should verify that dashboard content and refresh cadence meet their specific examination obligations.

Data Model

Connect to the following Dataverse tables. Use the OData connector in Power BI Desktop with your environment URL.

Table	Role
fsi_modelinventory	Primary — agent model inventory records
fsi_mrmriskrating	Risk scoring history per agent
fsi_validationcycle	Validation workflow lifecycle
fsi_validationfinding	Findings raised during validation
fsi_monitoringrecord	Scheduled performance monitoring data
fsi_mrmcomplianceevent	Audit trail of compliance-relevant events

Relationships

• fsi_modelinventory 1:N → fsi_mrmriskrating (on lookup column fsi_modelinventory_lookup; in Power BI, join on fsi_modelinventory_lookup_value)
• fsi_modelinventory 1:N → fsi_validationcycle (on lookup column fsi_modelinventory_lookup)
• fsi_validationcycle 1:N → fsi_validationfinding (on lookup column fsi_validationcycle_lookup)
• fsi_modelinventory 1:N → fsi_monitoringrecord (on lookup column fsi_modelinventory_lookup)
• fsi_modelinventory 1:N → fsi_mrmcomplianceevent (on lookup column fsi_modelinventory_lookup)

Note: Child tables reference the parent via Dataverse lookup columns (suffix _Lookup), not direct FK columns. The Dataverse Power BI connector exposes these as <lookupname>_value GUID columns. fsi_modelid is the alternate-key text on the parent (e.g., MRM-2026-00001); use it for display, not joins.

Dashboard Pages

Page 1: Inventory Overview

Visual	Type	Description
Total Agents in MRM Inventory	Card	COUNTROWS(fsi_modelinventory)
Agents Pending Submission	Card	Count where fsi_mrmstatus = 100000000 (Pending Submission). Note: Registered is the agent-registry-automation status, not an mrmstatus value.
Agents Validated	Card	Count where fsi_validationstatus = 100000006 (Validated). Use the integer value or join the _Label shadow column added by Power BI for the Choice type.
Agents by MRM Tier	Bar chart	X-axis: fsi_mrmtier (Tier 1–4), Y-axis: count
Risk Rating Distribution	Pie chart	Segments: Critical, High, Medium, Low from fsi_currentriskrating
Agent Inventory Detail	Table	Columns: Model ID, Name, Tier, Rating, Status, Next Validation Due

Filter bar: MRM Tier, Risk Rating, Validation Status, Business Function.

Page 2: Validation Status

Visual	Type	Description
Open Validation Cycles	Card	Count where fsi_cyclestatus not in (100000006 [Validated], 100000007 [Rejected])
Overdue Validations	Card	Uses Overdue Validations DAX measure (see below)
SLA Breaches	Card	Count of cycles where any SLA phase is breached
Validation Status by Tier	Stacked bar	X-axis: Tier, segments: cycle status values
Validation Cycle Durations	Timeline	Start: fsi_submitteddate, End: fsi_validationcompleteddate
Active Cycles Detail	Table	Cycle ID, Model Name, Phase, Assigned Validator, SLA Status, Days Remaining

Page 3: Findings & Remediation

Visual	Type	Description
Open Critical Findings	Card	Count where severity = Critical and status ≠ Closed
Total Open Findings	Card	Count where status ≠ Closed
Findings by Category	Bar chart	X-axis: fsi_findingcategory, Y-axis: count
Findings by Severity	Bar chart	X-axis: fsi_severity, Y-axis: count
Open Findings Detail	Table	Finding ID, Model Name, Category, Severity, Owner, Due Date, Status

Page 4: Monitoring Trends

Visual	Type	Description
Error Rate Trends	Line chart	X-axis: week, Y-axis: error rate, series: agent name
Escalation Rate Trends	Line chart	X-axis: week, Y-axis: escalation rate, series: agent name
Threshold Breaches This Month	Card	Count of monitoring records with breach flags
Recent Monitoring Alerts	Table	Agent Name, Metric, Threshold, Actual Value, Drift Signal, Date

Page 5: Compliance Events

Visual	Type	Description
Total Events (Last 30 Days)	Card	Filtered count of fsi_mrmcomplianceevent
Events by Type	Bar chart	X-axis: fsi_eventtype, Y-axis: count
Events by SR 11-7 Pillar	Pie chart	Segments from fsi_sr117pillar (option set fsi_mrm_sr117pillar: 100000000=Pillar 1, 100000001=Pillar 2, 100000002=Pillar 3)
Recent Compliance Events	Table	Event ID, Model Name, Event Type, Impact Level, Date, Details

Key DAX Measures

Agents Not Validated = 
CALCULATE(
    COUNTROWS(fsi_modelinventory),
    fsi_modelinventory[fsi_validationstatus] <> 100000006   -- 100000006 = Validated
)

Tier 1 Coverage = 
DIVIDE(
    CALCULATE(COUNTROWS(fsi_modelinventory), 
        fsi_modelinventory[fsi_mrmtier] = 100000001,   -- Tier 1
        fsi_modelinventory[fsi_validationstatus] = 100000006), -- Validated
    CALCULATE(COUNTROWS(fsi_modelinventory),
        fsi_modelinventory[fsi_mrmtier] = 100000001),
    0
)

Overdue Validations = 
CALCULATE(
    COUNTROWS(fsi_modelinventory),
    fsi_modelinventory[fsi_nextvalidationdue] < TODAY(),
    fsi_modelinventory[fsi_validationstatus] = 100000006   -- Validated
)

Avg Cycle Duration Days = 
AVERAGEX(
    FILTER(fsi_validationcycle, NOT(ISBLANK(fsi_validationcycle[fsi_validationcompleteddate]))),
    DATEDIFF(fsi_validationcycle[fsi_submitteddate], fsi_validationcycle[fsi_validationcompleteddate], DAY)
)

Open Critical Findings = 
CALCULATE(
    COUNTROWS(fsi_validationfinding),
    fsi_validationfinding[fsi_severity] = 100000001,   -- Critical (severity is 1-based)
    fsi_validationfinding[fsi_remediationstatus] <> 100000004   -- Closed (column is fsi_remediationstatus, not fsi_findingstatus)
)

Refresh Schedule

Setting	Recommended Value
Refresh frequency	Daily
Refresh time	06:00 AM UTC
Rationale	Flow 1 (Inventory Sync) runs at 05:00 AM UTC — scheduling the refresh one hour later helps capture the latest sync results

Implementation note: Configure scheduled refresh in the Power BI Service workspace settings after publishing. Gateway configuration may be required for on-premises Dataverse environments.

Row-Level Security (Optional)

For environments where agent owners should see only their own models:

1. Create a role named AgentOwnerFilter
2. Add DAX filter on fsi_modelinventory: [fsi_ownerupn] = USERPRINCIPALNAME()
3. Assign agent owners to this role in Power BI Service
4. MRM team and examiners should not be assigned to this role

Publication Checklist

• All five pages render with live data
• DAX measures return expected values
• Refresh schedule configured (daily at 06:00 AM UTC)
• Row-level security configured (if applicable)
• Dashboard shared with MRM team and governance stakeholders
• Document Power BI workspace URL in DELIVERY-CHECKLIST.md