WGU D605 | OPTIMIZATION
Task 2
Amazon Air Fleet Route
Optimization Analysis
A structured business optimization assessment covering the problem definition, mixed-integer model, algorithm
selection, technology stack, operational risks, and a reproducible planning example.
Course D605 - Optimization
Assessment Task 2 - Performance Assessment
Business context Amazon Air - air cargo route and fleet planning
Analysis date October 2026
Evidence boundary Scenario-based report with independently verified synthetic example
Scope and evidence note. The uploaded Task 2 document supplies the A-E section framework and Amazon Air case
description. WGU's October 2026 catalog independently confirms D605 course topics but does not disclose the current Task 2
rubric. All numerical route data, fleet capacities and dollar costs in the worked demonstration are explicitly hypothetical and
are not Amazon operating records.
WGU D605 | Task 2 | Amazon Air Fleet Route Optimization 1
, Executive Summary
Amazon Air operates within an air logistics network where cargo demand, aircraft availability, route sequencing,
turnaround times, and delivery deadlines must be coordinated. Amazon publicly identifies the
Cincinnati/Northern Kentucky International Airport (CVG) operation as a major air hub; its August 2026
announcement reported more than three dozen daily flights at that site [2]. This report uses that verified
business context, not confidential company schedules, to examine a routing decision.
The decision is to select aircraft and service sequences that minimize a combined cost of flying, aircraft
activation, cargo handling, and late delivery while delivering every modeled shipment, respecting payload
capacities, avoiding disconnected routes, and returning aircraft to their origin within a scheduling period. The
underlying class is a capacitated, time-window-constrained vehicle routing problem with a heterogeneous fleet.
Its route choices are discrete and the arrival and delay decisions are continuous, making mixed-integer linear
programming (MILP) appropriate [3, 4].
A comprehensive arc-based MILP formulation is given in Section B. A small equivalent route-enumeration MILP is
solved in Section C to provide verifiable numerical evidence. It examines 38 feasible aircraft-specific tours
and achieves an optimal hypothetical total cost of $8,505, compared with $8,810 for an explicitly defined
feasible manual plan, a 3.46% improvement. The solution uses two of three available aircraft, satisfies all four
modeled destination demands, incurs no lateness, and attains a zero reported MIP gap. These results show the
method can be validated; they do not establish actual cost reductions at Amazon.
The recommended implementation is a staged decision-support process: validate route and shipment data; solve
with an exact MILP solver; check all constraints independently; compare with a current baseline; stress-test
disruptions; and permit dispatch staff to accept or override a plan. For production use, flight feasibility, operator
certifications, airport access, payload restrictions, crew duty limits, real-world time zones, and schedule recovery
must be incorporated before any candidate plan could be operationally approved.
Assessment-structure crosswalk
Uploaded Task 2 section Where covered here
A. Optimization Problem Identification A: problem, scope, business requirement, classification
B1. Objective Function B1: costs, index sets, objective function
B2. Constraint Expressions B2: service, capacity, continuity, time, return and domain constraints
B3. Decision Variables B3: four core classes plus visit indicators
C1. Method and Algorithm C1: MILP, branch-and-bound / branch-and-cut and alternatives
C2. Tools and Technologies C2: Python, solver, data, verification and deployment
D. Risks and Limitations D: operational, technical, ethical and model risks
E. References E: traceable primary, software and scholarly sources
Alignment boundary: These headings are grounded in the uploaded document; they are not represented as an
independently verified current WGU scoring rubric. The WGU catalog establishes only the course-level scope [1].
WGU D605 | Task 2 | Amazon Air Fleet Route Optimization 2
Task 2
Amazon Air Fleet Route
Optimization Analysis
A structured business optimization assessment covering the problem definition, mixed-integer model, algorithm
selection, technology stack, operational risks, and a reproducible planning example.
Course D605 - Optimization
Assessment Task 2 - Performance Assessment
Business context Amazon Air - air cargo route and fleet planning
Analysis date October 2026
Evidence boundary Scenario-based report with independently verified synthetic example
Scope and evidence note. The uploaded Task 2 document supplies the A-E section framework and Amazon Air case
description. WGU's October 2026 catalog independently confirms D605 course topics but does not disclose the current Task 2
rubric. All numerical route data, fleet capacities and dollar costs in the worked demonstration are explicitly hypothetical and
are not Amazon operating records.
WGU D605 | Task 2 | Amazon Air Fleet Route Optimization 1
, Executive Summary
Amazon Air operates within an air logistics network where cargo demand, aircraft availability, route sequencing,
turnaround times, and delivery deadlines must be coordinated. Amazon publicly identifies the
Cincinnati/Northern Kentucky International Airport (CVG) operation as a major air hub; its August 2026
announcement reported more than three dozen daily flights at that site [2]. This report uses that verified
business context, not confidential company schedules, to examine a routing decision.
The decision is to select aircraft and service sequences that minimize a combined cost of flying, aircraft
activation, cargo handling, and late delivery while delivering every modeled shipment, respecting payload
capacities, avoiding disconnected routes, and returning aircraft to their origin within a scheduling period. The
underlying class is a capacitated, time-window-constrained vehicle routing problem with a heterogeneous fleet.
Its route choices are discrete and the arrival and delay decisions are continuous, making mixed-integer linear
programming (MILP) appropriate [3, 4].
A comprehensive arc-based MILP formulation is given in Section B. A small equivalent route-enumeration MILP is
solved in Section C to provide verifiable numerical evidence. It examines 38 feasible aircraft-specific tours
and achieves an optimal hypothetical total cost of $8,505, compared with $8,810 for an explicitly defined
feasible manual plan, a 3.46% improvement. The solution uses two of three available aircraft, satisfies all four
modeled destination demands, incurs no lateness, and attains a zero reported MIP gap. These results show the
method can be validated; they do not establish actual cost reductions at Amazon.
The recommended implementation is a staged decision-support process: validate route and shipment data; solve
with an exact MILP solver; check all constraints independently; compare with a current baseline; stress-test
disruptions; and permit dispatch staff to accept or override a plan. For production use, flight feasibility, operator
certifications, airport access, payload restrictions, crew duty limits, real-world time zones, and schedule recovery
must be incorporated before any candidate plan could be operationally approved.
Assessment-structure crosswalk
Uploaded Task 2 section Where covered here
A. Optimization Problem Identification A: problem, scope, business requirement, classification
B1. Objective Function B1: costs, index sets, objective function
B2. Constraint Expressions B2: service, capacity, continuity, time, return and domain constraints
B3. Decision Variables B3: four core classes plus visit indicators
C1. Method and Algorithm C1: MILP, branch-and-bound / branch-and-cut and alternatives
C2. Tools and Technologies C2: Python, solver, data, verification and deployment
D. Risks and Limitations D: operational, technical, ethical and model risks
E. References E: traceable primary, software and scholarly sources
Alignment boundary: These headings are grounded in the uploaded document; they are not represented as an
independently verified current WGU scoring rubric. The WGU catalog establishes only the course-level scope [1].
WGU D605 | Task 2 | Amazon Air Fleet Route Optimization 2
