ViaViz Logistics Lab

Logistics Lab / edge workbench

Inspect routing behavior
across space and time.

ViaViz is a visual workbench for replaying and inspecting routing solutions in dynamic multi-agent logistics systems.

It helps researchers, students, and practitioners examine how vehicles, vessels, drones, robots, human agents, and other entities move and interact across logistics networks.

Open ViaViz Follow the tutorial

Visual evidence complements numerical evidence. Distance, waiting time, service rate, utilization, and rejected requests remain essential; replay adds the spatial-temporal account behind them.

Import files Explore benchmarks Connect a simulator

Operations atlas / live replays

Four routing contexts. One inspection lens.

Select a replay or open the complete workspace.

Why visual inspection?

Aggregate indicators do not always explain operational behavior.

Visual replay can expose concentration, cycling, unexpected detours, waiting, delayed service, and other patterns that deserve closer examination.

01

Debug

Trace route and stop sequences, time windows, service events, loads, and network geometry when a solution behaves unexpectedly.

02

Validate

Check whether movement, waiting, dispatch, rebalancing, and request handling agree with the intended model and operating rules.

03

Teach and communicate

Use a replay to discuss how a routing solution unfolds while keeping the underlying quantitative evaluation in view.

Inspection scope

What can be inspected?

Detailed tutorial

From routing output to an inspectable replay.

Begin with a bundled scenario, a literature instance, or compatible route and request data from your own computational workflow.

  1. 01

    Open a representative case

    Start with one of the four live examples above, or choose Literature library to inspect an available benchmark instance and solution.

    Try Solomon C101
  2. 02

    Control the replay horizon

    Use the status-bar timeline to choose the analysis interval, scrub the current time, pause playback, and study a specific operational moment.

  3. 03

    Inspect agents and requests

    Select a vehicle or request to reveal its route, paired endpoints, load, timings, and current state. Open the fleet, demand, and time-window charts for supporting context.

  4. 04

    Compare the replay with the indicators

    Relate patterns on the network to waiting, service, delay, rejection, distance, and utilization measures. Treat visual findings as questions to verify numerically.

  5. 05

    Capture a reproducible view

    When a compatible service backend is available, use Share view to preserve the scene, camera, selected horizon, playback time, and workspace settings in a revocable URL.

Use your own output

Inspect your own routing solution

ViaViz uses a documented data contract for route and request information. Compatible output can come from Python, Java, Julia, C++, optimization solvers, simulation models, or other computational workflows.

The local bridge can open selected solver outputs in the hosted workbench while serving those selected files from the user’s machine. Its scope and security depend on the configured deployment and browser environment.

Load files Open connector tutorial
Example payload aviz-result.v1 
{
  "schema_version": "aviz-result.v1",
  "title": "Routing experiment",
  "routes": [
    { "id": "vehicle-01", "path": { "type": "FeatureCollection", "features": [] } }
  ]
}

Preserve the facts your solver produced. Add geometry, demand, timing, and metadata when those values are available.

Background

Developed through research and teaching.

Research origin

Why ViaViz?

During Breno’s PhD work on mobility-on-demand systems, visual replay helped explain behavior that aggregate KPIs did not describe adequately. Vehicles could concentrate excessively in high-demand areas, ignore diminishing returns, or move cyclically between attractive regions.

The lesson was not that the KPIs were wrong. Aggregate indicators alone did not always explain how a policy behaved in space and time.

Teaching use

A workbench used with routing algorithms

A version of this visualization approach was later used in the Transportation and Logistics Management course. Students could test routing algorithms, inspect generated routes, and validate whether their solutions behaved as intended.

ViaViz remains an early research and teaching prototype. This account does not imply a controlled educational-effect study.

Logistics Lab

An edge workbench from Logistics Lab

Logistics Lab is a home for ideas, tutorials, benchmarks, experiments, and software related to dynamic multi-agent logistics systems. ViaViz focuses on the edge side of supply-chain networks: movement, routing, dispatch, and service behavior over time.

Future workbenches may examine operations at network nodes such as warehouses, depots, and terminals.

Via refers to a way, road, route, or path in Latin. Viz refers to visualization. The name describes visual inspection of what happens along the routes and edges of logistics networks.

Current research prototype

Explore a scenario and inspect what the indicators cannot show alone.

Test the current prototype and share technical feedback.

Open ViaViz Python connector tutorial