Aurora Banking Group · Banking

Banking Workforce Digital Twin: Simulate Before You Commit

Maturity
L5
Domain
Plan & Cost
Analytics
prescriptive
Scenario reshaping feasibility
2 of 3
Sponsor
Group Chief Operating Officer
Confidence
High

The situation

Under a given AI-adoption, automation, agility and demand scenario, what workforce — by role, skill and control coverage — does Aurora need, can it get there in time, and does it remain a controlled bank throughout?

Plan & CostL5Sponsor · Sofia Almeida

The recommendation on the table

Make BNK-05 a required gate for major operating-model and AI-capital decisions

Turns the most expensive transformation decisions from leaps of faith into sequenced, governed, simulated choices.

Decision ownerGroup Chief Operating Officer · Sofia Almeida
MaturityL5
Priorityhigh

Trade-offAdds a step to capital approval and depends on the upstream projects' data staying current.

The evidence

Aurora's most expensive workforce decisions — operating-model redesigns, AI-capital commitments, segment restructurings — were made with assumed, not simulated, workforce consequences. The integrative capstone of the Banking portfolio, BNK-05 consumes the BNK-01 baseline, BNK-02 productivity and BNK-03 exposure to simulate the bank's workforce under transformation scenarios before the capital is committed — including, distinctively, whether the bank stays controlled as roles automate. It found the aggressive scenario infeasible on both reskilling pace and control coverage, the base scenario feasible and resilient, and control coverage — not value — the binding constraint on the fastest paths.

Aurora — Banking Workforce Digital Twin

Compare transformation scenarios — feasibility, resilience and whether the bank stays controlled — before the capital is committed.

Scenario reshaping feasibility· transformation scenarios feasible within timeline
2 of 3
Critical
Control coverage under automation· aggressive scenario drops coverage below threshold mid-transition
breaches
Critical
Workforce resilience index· composite coverage/flex/SPOF under stress (0–1)
0.72
On watch
Reskilling throughput vs. need· aggressive scenario needs ~1.8x current throughput
-45%
Critical
Illustrative preview
Scenario reshaping feasibility (margin, quarters)
01.252.53.755CautiousBaseAggressive

Key takeawayThe aggressive scenario is infeasible on reskilling pace; the base scenario holds with margin.

Workforce resilience under stress (index, 0–100)
0255075100CautiousBaseAggressive

Key takeawayThe base scenario is resilient; the aggressive scenario is fragile under attrition stress.

Interactive view is best explored on desktop.

Key findings

Run through the twin, the aggressive AI-adoption scenario breaks on two constraints at once: it needs about 1.8× current reskilling throughput, and it drops control coverage below threshold mid-transition. The base scenario, by contrast, is feasible and resilient. The fastest path is not the best path.

2 of 3of transformation scenarios feasible in time

What we can’t claim

The twin shows that the highest-value scenarios are the ones most likely to leave the bank under-controlled as roles automate. A path that maximises value but drops control coverage below threshold is not a high-value path — it is a regulatory failure waiting to happen, and the twin is what makes that visible before the capital is committed.

Recommendations

Make BNK-05 a required gate for major operating-model and AI-capital decisions

high priority

Turns the most expensive transformation decisions from leaps of faith into sequenced, governed, simulated choices.

Trade-off

Adds a step to capital approval and depends on the upstream projects' data staying current.

Commit the base scenario; stage the aggressive path; treat control coverage as a hard constraint

high priority

Captures most of the value on a feasible, resilient path while refusing any scenario that de-controls the bank.

Trade-off

Leaves some theoretical value unbooked in exchange for feasibility and control — the right trade for a regulated bank.

Analytical framework

How we reached this

Prescriptive simulation — model the workforce consequence of transformation scenarios (feasibility, value, control) before capital is committed.

ConfidenceMedium

Methods applied

Scenario simulation (digital twin)Monte Carlo simulationWorkforce allocation optimizationDecision modelingControl-coverage simulation

Statistical techniques

Scenario simulationOptimizationRisk scoringForecasting

Algorithms

Monte Carlo simulationLinear programmingNetwork analysisTime-series forecasting

Data sources

BNK-01 baselineBNK-02 productivityBNK-03 exposure & reshapingControl-obligation registerSkills inventoryMobility data

Outputs generated

Scenario feasibility with marginControl-coverage trajectoriesWorkforce resilience indexRecommended reshaping path

Why this confidence

The most advanced method in the portfolio, yet bounded by assumption load; outputs are always ranges against stated assumptions — sophistication and confidence are independent.

The reasoning

Business context

The most advanced Banking project, consuming all upstream Banking work. Sponsored by the Group COO with the Transformation office, because its outputs gate the bank's largest operating-model and AI-capital decisions.

Expected value

De-risks the most expensive transformation decisions by simulating their workforce and control consequences before commitment, and avoids the mis-sequenced transformations and control failures that overrun on workforce, not technology.

Workforce landscape

Under demand and AI-capability stress, the aggressive scenario breaks on reskilling pace and drops control coverage below threshold mid-transition; the base scenario holds. Resilience is thinner than steady-state metrics suggest, and certification of new governance capability is a binding lead-time.

The analytics journey

Level 5, prescriptive. Simulation and optimisation: it compares scenarios rather than predicting one future, and every output carries its assumptions and a range. Explicit that a scenario is a decision aid, not a forecast, and only as good as its inputs — which is why it depends on the three upstream projects.

Under the hood

A scenario engine computes required workforce by role, skill and control coverage as a function of AI-adoption, automation, agility and demand assumptions; Monte Carlo produces outcome distributions; linear programming finds the feasible reshaping path; network analysis surfaces control single-points-of-failure. A scenario that maximises value but drops control coverage below threshold is flagged infeasible regardless of its value.

Confidence & evidence

Why you can rely on this

86%
Analysis confidenceHigh

The inconvenient truth

The twin shows that the highest-value scenarios are the ones most likely to leave the bank under-controlled as roles automate. A path that maximises value but drops control coverage below threshold is not a high-value path — it is a regulatory failure waiting to happen, and the twin is what makes that visible before the capital is committed.

Method

Confidence is a deterministic read of KPI strength, target and benchmark coverage across this project — shown on an illustrative reference dataset, computed the same way it would be on live data.

Take this further

Where this project connects