Last-Mile Delivery Research: Analysing Urban Logistics Challenges

The last mile is where supply chains meet customers — and where costs, complexity and customer expectations collide. Research Bureau delivers evidence-based, actionable research that helps logistics leaders, city planners and retailers solve urban last-mile challenges and unlock measurable savings, speed and sustainability gains.

This page explains how we approach last-mile research, the core urban logistics problems we analyse, methodologies and deliverables, real-world examples, and how to start a tailored engagement. Share project details for a customised quote via our contact form, the WhatsApp icon on this page, or email us at [email protected].

Why last-mile research matters now

Last-mile operations typically consume 28–53% of total delivery costs and account for a disproportionate share of urban congestion and emissions. Optimising the last mile is therefore a high-impact opportunity for cost reduction, customer experience improvement and regulatory compliance.

  • Commercial impact: Reduced variable cost per delivery, fewer failed deliveries, higher first-time success rates and optimised fleet utilisation.
  • Operational impact: Smarter routing, reduced dwell times, better use of micro-hub and locker networks, and improved labour planning.
  • Strategic impact: Scalable delivery models that align with sustainability targets and city regulations while improving customer retention.

Key urban logistics challenges we analyse

Urban environments introduce a unique set of constraints and trade-offs. Our research examines these pain points in depth:

  • Traffic congestion and route unpredictability
  • High population density versus limited curbside space
  • Loading and parking constraints for different vehicle types
  • Local regulations, low-emission zones and curfew restrictions
  • Environmental targets and emissions accounting
  • Last-mile data fragmentation and poor telemetry
  • Fleet heterogeneity and asset underutilisation
  • Labour shortages, wages and compliance risks
  • Rising consumer expectations for speed and delivery windows
  • Reverse logistics and management of returns
  • Security risks: theft, vandalism and package loss

Each challenge is assessed quantitatively and qualitatively to surface root causes and target interventions with the highest ROI.

Deep-dive: congestion, curb space and the parking problem

Traffic congestion and limited loading zones are among the most persistent constraints in dense cities. We model curbside demand, dwell-time distributions and illegal parking impacts to quantify lost productivity and enforcement costs.

  • We use high-resolution GPS and curb usage datasets to map peak demand cycles.
  • Simulation models evaluate the effect of various interventions (dedicated loading bays, dynamic pricing for curb use, micro-hub deployment).
  • The outcome includes policy suggestions, design specifications for loading infrastructure and a cost-benefit analysis for private and public stakeholders.

Example outcome: a combined strategy of scheduled delivery windows and micro-hub adoption can reduce average dwell time by 18–35% and increase delivery throughput per curb space by up to 40% in trials we design and oversee.

Fleet strategy and mode choice: choosing between vans, bikes, lockers and drones

Selecting the right delivery modes for different urban contexts directly affects cost, emissions and service levels. We evaluate mode choice across density bands and parcel profiles.

Mode Typical cost per delivery Speed (urban core) Emissions footprint Best-suited context Regulatory complexity
Sprinter van Medium–high Medium High Medium-density suburban routes Moderate
Small van / LCV Medium High Medium Suburban & peri-urban Low–moderate
Cargo bike Low High in dense cores Low High-density downtown Low
E-bike/e-cargo Low High Very low Short-radius dense deliveries Low
Parcel lockers Very low per pick-up Variable Low High footfall nodes Low–moderate
Drones (pilot stage) High Very high Low Time-critical, low-obstruction areas High

We pair this analysis with a routing cost model that computes blended costs and service levels across thousands of scenarios, enabling decisions like when to shift high-volume segments to micro-hubs serviced by cargo bikes.

Data integration and telemetry: turning fragmentation into insights

Fragmented data (multiple carriers, different TMS/EMS formats, siloed telemetry) prevents reliable optimisation. Our approach consolidates and harmonises datasets for usable analytics.

  • We implement ETL pipelines to ingest GPS traces, POD timestamps, order metadata and IoT sensor data.
  • Data validation checks mitigate missing telemetry and reconcile multi-actor event timelines.
  • The result is a single source of truth for KPI calculation, heatmapping and anomaly detection.

Deliverables include cleaned datasets, a reproducible analytics pipeline and a KPI dashboard that stakeholders can use for ongoing decision-making.

Route optimisation vs. reality: accounting for urban variability

Classic route optimisation often fails when it ignores urban variability: temporary roadworks, parades, school pick-ups and illegal stops. We overlay stochastic elements into routing simulations.

  • We use probabilistic travel time distributions rather than deterministic averages.
  • Monte Carlo simulations test route robustness across thousands of scenarios.
  • Route plans are stress-tested for worst-case peak congestion and validated against historical GPS traces.

This leads to robust routing policies that balance efficiency with predictability and SLA adherence.

Case studies and illustrative examples

Note: the following case studies are anonymised client engagements to illustrate typical outcomes and methodologies.

Case study A — Metropolitan grocery retailer

  • Challenge: High failed delivery rates in inner city, rising cost per delivery.
  • Research approach: Zip-code level demand clustering, pilot of e-cargo bikes from newly established micro-hubs.
  • Outcome: 26% reduction in urban delivery costs for inner-city zones and a 15% increase in first-time delivery success. Operational model scaled across 3 cities after pilot validation.

Case study B — National e-commerce platform

  • Challenge: Complex returns flows and lack of visibility across multiple carrier partners.
  • Research approach: Integrated telemetry, reverse-logistics optimization and locker network expansion analysis.
  • Outcome: Consolidated returns routing reduced average reverse-logistics cost by 22% and improved customer returns cycle time by 40%.

Case study C — City transport authority

  • Challenge: Rising congestion and conflicting curb use between ride-hailing and delivery fleets.
  • Research approach: Curb-side demand mapping, stakeholder workshops and a pilot shared-loading bay scheme.
  • Outcome: Pilot reduced illegal loading incidents by 47% and increased delivery throughput during off-peak hours, supporting a permanent curb allocation policy.

Our methodology — rigorous, reproducible, stakeholder-led

We combine field data, quantitative modelling and collaborative stakeholder engagement to produce research that drives implementation.

  • Diagnostic phase: Stakeholder interviews, existing data audit, hypothesis generation.
  • Data collection: GPS traces, order records, driver diaries, curbside observations, API integrations.
  • Modelling and simulation: Cost models, route optimisation, Monte Carlo risk analysis, GIS-based heatmaps.
  • Pilot design: Controlled trials, KPIs, A/B comparisons and user acceptance testing.
  • Policy and implementation planning: Regulatory impact assessment, procurement recommendations and rollout roadmap.

Each engagement is documented with reproducible code and data pipelines where permitted, ensuring transparency and long-term maintainability.

KPIs we measure and optimise

We focus on operational and strategic KPIs that matter to executives, operations and city planners.

  • Cost per delivery (fully loaded)
  • Cost per successful first-time delivery
  • Failed delivery rate and recovery cost
  • Average dwell time at delivery stop
  • Door-to-door delivery time and SLA compliance
  • CO2e per parcel and local NOx/PM impacts
  • Utilisation rates per vehicle and per driver
  • Locker pick-up rates and locker utilisation
  • Reverse-logistics cost and return lead time
  • Customer satisfaction (CSAT) related to delivery

These KPIs form the basis of scenario analysis and ROI estimation.

Technology stack and analytical tools

We deploy a blend of off-the-shelf and custom analytical tools depending on client needs and data sensitivity.

  • GIS platforms for spatial analysis and heatmaps
  • Advanced routing engines and optimisation libraries
  • Time-series and telemetry analytics platforms
  • SQL/noSQL databases and reproducible ETL pipelines
  • Dashboards (Power BI, Tableau or client-preferred) for ongoing monitoring

We advise on architecture choices and integration patterns to ensure research outputs are operationally actionable.

Deliverables you can act on

Our research engagements culminate in a package of deliverables tailored to decision-making and implementation.

  • Executive summary and strategic recommendations
  • Detailed technical report with methodologies and assumptions
  • Cost-benefit analysis and ROI scenarios
  • GIS heatmaps and demand clusters
  • Optimised route templates and fleet-mix recommendations
  • Pilot design, KPIs and evaluation plan
  • Implementation roadmap and procurement advice
  • Data pipelines, cleaned datasets and code notebooks (where permitted)
  • Presentation and stakeholder workshop to drive adoption

Each deliverable is explained in plain language for executive audiences and accompanied by technical annexes for operations teams.

Typical project timelines and scope options

Project timelines depend on scale, data availability and pilot complexity. Typical engagement sizes include:

  • Rapid diagnostic (2–4 weeks): High-level assessment, quick wins and an initial roadmap.
  • Full urban last-mile study (8–12 weeks): Comprehensive data collection, modelling, and recommendations.
  • Pilot design and support (12–24 weeks): Pilot deployment, data collection and iterative optimisation.
  • Ongoing partnership and monitoring (retainer): Continuous improvement and dashboarding.

We customise timelines based on your goals, resource availability and regulatory window.

Pricing models and commercial terms

We offer flexible engagement models to match client needs and risk appetite.

  • Fixed-fee scoping and delivery: Clear deliverables and timelines, ideal for defined outcomes.
  • Time & materials: Agile engagements where scope evolves.
  • Pilot co-funding: Shared risk models for pilots with capital expenditures.
  • Retainer and advisory: Ongoing monitoring and rapid response capability.

Share project context and constraints for a tailored quote. We can sign NDAs prior to data sharing and align commercial terms with expected ROI.

Stakeholder engagement and governance

Last-mile solutions require multi-stakeholder collaboration. We design governance frameworks that bring carriers, city agencies, retailers and community groups together.

  • Stakeholder mapping and engagement plans to surface constraints and incentives.
  • Transparent data-sharing agreements and privacy-compliant telemetry handling.
  • Workshops to validate model assumptions and align KPIs across actors.
  • Change-management support to ensure pilots scale into operations.

This ensures research translates into implementable solutions with stakeholder buy-in.

Policy, regulation and sustainability considerations

We evaluate regulatory constraints (low-emission zones, noise curfews), local policy levers (curb-management pricing) and sustainability targets (Scope 3 emissions). Our research produces policy-ready recommendations that balance operational feasibility with environmental outcomes.

  • Emissions accounting aligned to GHG Protocol and local reporting needs.
  • Cost and operational impacts of low-emission zone policies modelled across scenarios.
  • Recommendations for electrification rollouts, including charging infrastructure and depot location analysis.

These outputs help clients meet compliance and corporate sustainability goals without sacrificing service.

Implementation risk management

Scaling last-mile changes carries operational risk. We identify and quantify risks and propose mitigations.

  • Supplier risk: multi-carrier strategies and fallback routing.
  • Operational risk: phased rollouts and rigorous KPIs for early detection.
  • Regulatory risk: scenario planning and advocacy playbooks.
  • Financial risk: staged investments tied to pilot performance milestones.

Our research includes contingency plans and go/stop decision criteria to keep projects on track.

Frequently asked questions

  • What data do you need to start?
    We typically need order and routing data, carrier logs, fleet characteristics and a high-level business brief. We can work with partial data and recommend data collection improvements.

  • How long before we see actionable results?
    Rapid diagnostics can surface quick wins within weeks. Full optimisation and pilot validation typically require 8–12 weeks.

  • Do you run pilots and proof-of-concepts?
    Yes. We design, manage and analyse pilots to validate models and estimate real-world impact.

  • Can you work with public sector clients?
    Yes. We work with municipal authorities, transport planners and public-private partnerships to design solutions that balance public interest and commercial viability.

  • How do you handle sensitive data?
    We sign NDAs and follow secure data handling practices, encrypting sensitive datasets and restricting access on a need-to-know basis.

Why choose Research Bureau

Research Bureau brings a pragmatic, evidence-first approach to last-mile research. Our strengths include:

  • A cross-disciplinary team of senior supply-chain researchers, GIS specialists and logistics analysts with decades of combined experience.
  • Proven methodologies that combine field observation, telemetry, GIS analysis and optimisation modelling.
  • Focus on actionable deliverables and measurable ROI, not just academic reports.
  • Flexible commercial models and the ability to support pilots through to scale.

We prioritise transparency, reproducibility and stakeholder alignment to ensure research leads to real operational change.

How to engage us — simple steps to a bespoke quote

To get started and receive a tailored proposal, share a brief description of your objectives. We’ll respond with a scoped next step and price estimate.

  • Step 1: Share your project brief or request a discovery call via the contact form.
  • Step 2: We conduct a rapid scoping workshop and outline data needs.
  • Step 3: You receive a proposal with timelines, costs and a pilot/design plan.
  • Step 4: Upon approval, we begin the diagnostic phase and schedule stakeholder sessions.

Click the WhatsApp icon on this page to start a direct chat, use the contact form to upload documents, or email [email protected] to provide details and request a quote.

Example deliverable: sample findings snapshot

Metric Baseline Intervention Projected impact
Average dwell time 6.2 minutes Micro-hub + scheduled windows 4.1 minutes (-34%)
Cost per delivery ZAR 58.50 Fleet optimisation & cargo bike shift ZAR 44.60 (-24%)
Failed delivery rate 12% Smart ETAs and alternative pick-up 6% (-50%)
CO2e per parcel 1.8 kg Electrified micro-hub + cargo bikes 1.1 kg (-39%)

These figures are illustrative; actual results depend on local conditions, scale and data quality. We provide tailored projections based on your data during scoping.

Common interventions we recommend (and when they work best)

  • Micro-hubs + cargo bikes — best for dense urban cores with high parcel volumes and limited curb access.
  • Parcel locker networks — effective where footfall is high and customers are willing to pick up at convenience nodes.
  • Dynamic curb management — suitable for cities with high dispute over curb space; needs municipal cooperation.
  • Scheduled delivery windows & customer UX changes — low-cost intervention to reduce dwell time and failed deliveries.
  • Electrification + depot charging — best where emissions targets are binding and fleet renewal is planned.

Each recommendation is paired with cost estimates, operational adjustments and pilot criteria.

Long-term value and ROI

Investing in last-mile research creates compounding value through:

  • Lower ongoing operating costs from improved routing and fleet utilisation.
  • Faster delivery cycles that increase customer lifetime value and reduce churn.
  • Lower environmental compliance risk and improved brand reputation.
  • Scalable solutions that can be replicated across regions.

We quantify these benefits in financial terms in our proposals so leadership can make data-driven investment decisions.

Ready to start? Get a tailored quote

Provide a short brief about your organisation, project goals, geographic scope and key constraints via the contact form, click the WhatsApp icon for an immediate chat, or email us at [email protected]. We will review your inputs and propose a customised scope and estimate within 48 hours.

We welcome clients from e-commerce platforms, retail chains, logistics providers, municipal agencies and technology vendors. Share your timeline and any immediate pain points to prioritise quick wins during the study.

Research Bureau — Evidence-driven supply chain and logistics research that turns urban last-mile complexity into strategic advantage. Contact us today to convert last-mile challenges into measurable outcomes.