Selected work · Applied analytics

Applied Analytics Casework

Selected analytical casework spanning preference research, predictive modeling, optimization, recommendation systems, and simulation.

Analytical casework

The challenge

A model matters only when it sharpens the decision.

Leaders often have a consequential question but not a dependable way to connect fragmented data, uncertain assumptions, and competing tradeoffs to a decision they can defend.

The response

Use the method the question requires.

This selected body of analytical work informs Axlete’s approach: structure the decision, choose fit-for-purpose methods, compare alternatives, and make the decision artifact inspectable. The public atlas preserves those decision structures. The separate capacity exercise below uses a constructed scenario data, not an original client result.

Decision modelingMachine learningOptimization & simulation

Casework atlas

The question is only the beginning.

Select a case to inspect the evidence structure, analytical method, and decision artifact. Nothing advances until the visitor asks the next question.

Atlas · AA-01Six decision structuresSanitized public casework

Preference · configuration · simulation

See which tradeoffs can sustain a coherent offer.

Preference work connects what people choose to the configuration decisions a product team can actually make. Competing offers remain visible long enough to compare tradeoffs rather than collapsing too early into one answer.

01 · Question
Which combination of safety, autonomy, connectivity, energy, and comfort should shape a new mobility offer?
02 · Evidence structure
Attribute levels, respondent choices, segment definitions, and candidate offer configurations.
03 · Method
- Conjoint analysis
- Customer segmentation
- Choice simulation
04 · Decision artifact
Offer-configuration comparison with a simulated market-choice readout.

Evidence boundary

No respondent records, client products, proprietary utility weights, or market-share estimates are shown. The public view demonstrates the decision structure—not a private result.

Public edition

Applied analytics · Inspectable numerical exercise

The constraint changes the answer.

Explore one capacity decision behind the atlas. This constructed numerical exercise demonstrates the method; it is not a result from the original casework.

Interactive example · Constructed data · Not a client engagementCapacity fixture CA-01

Available hours: 24ObjectiveMaximum demand for B: 8

6 A units. 3 B units.

Selected from every feasible whole-unit allocation under the capacity and demand constraints. The objective is the largest contribution.

Contribution$690

Completed units9

Hours used / available24 / 24

Unused hours0

Inspect the fixture, objective & calculation

Constructed scenario dated October 1, 2026. Product A: 3 hours/unit, $90 contribution/unit, demand cap 6. Product B: 2 hours/unit, $50 contribution/unit, demand cap 8. Both quantities must be nonnegative integers.

Constraint: 3A + 2B ≤ 24; 0 ≤ A ≤ 6; 0 ≤ B ≤ 8. Contribution = 90A + 50B. Completed units = A + B. The model enumerates all feasible pairs; ties in unit volume prefer higher contribution. Remaining ties retain the first enumerated pair.

Selected calculation: 90 × 6 + 50 × 3 = $690. No setup costs, sequencing, uncertainty, demand estimation, inventory, or real client data are included. Contribution is not revenue or net profit.

Decision artifact: a feasible allocation with its objective, constraints, unused capacity, and assumptions attached.

Decision Brief

A clear decision starts with a bounded question.

A written brief, source register, explicit assumptions, recommendations, and supporting analysis agreed in scope.

Indicative: 10 business days after the question, scope, and source access are agreed. Fixed fee agreed before work begins.

Request the brief