MATRA
MATRA is a Python-based freight transport and logistics simulation model, developed to investigate the operational implications of road-freight electrification.
It can be classified as an agent-based transport model, although its scope differs from general-purpose transport models such as MATSim.
What it represents
MATRA represents five components of a logistics system as interacting agents:
| Component |
|---|
| Freight depots |
| Freight vehicles |
| Customers |
| Public charging infrastructure |
| Electricity substations |
That last one is the distinguishing feature. MATRA models the electricity network alongside the transport network, which is what allows it to ask whether a charging strategy is actually deliverable.
How it works
The model combines three methods:
| Method | Contribution |
|---|---|
| Agent-based simulation | The behaviour of depots, vehicles and customers |
| Road-network routing | Movement over real road geometry |
| Mathematical optimisation | Efficient vehicle routing and charging decisions |
Between them these reproduce:
- Freight demand
- Vehicle allocation
- Delivery routing
- Vehicle energy consumption
- Charging decisions
- Charging-infrastructure utilisation
- Electricity-network constraints
Routing
MATRA represents freight movements over a digital road network derived from OpenStreetMap data.
Freight locations are connected to the road network, and road distances between relevant depots, customers and charging locations are calculated using shortest-path routing. Those distances are then incorporated into the operational decision-making of logistics agents, including vehicle-routing and charging optimisation.
Fleet configurations
The model can simulate alternative fleet configurations:
- Conventional internal-combustion vehicles
- Electric vehicles
- Mixed fleets
It can therefore be used to investigate the effects of freight electrification and alternative charging-infrastructure strategies.
Important limitations
MATRA is not a network traffic-assignment model
The handbook is explicit that the current implementation should be distinguished from a complete network traffic-assignment model. Three limitations follow, and they matter when interpreting results:
Network conditions are not endogenous. Road-network conditions are not generated through interactions between MATRA freight vehicles and the wider population of road users.
Travel time is assumed, not simulated. It is currently derived from road distance and an assumed average freight-vehicle speed, rather than from dynamically simulated congestion.
Other modes are absent. Interactions with passenger cars, public transport, walking, cycling and other transport modes are not explicitly represented.
The practical consequence is a matter of scope rather than quality. MATRA focuses on the operational behaviour and infrastructure requirements of the freight and logistics system, rather than reproducing the complete multimodal transport system.
Within the DARe modelling framework it is therefore best regarded as a specialised freight transport model that complements broader transport-system models such as MATSim.
How MATRA and MATSim divide the work
| MATSim | MATRA | |
|---|---|---|
| Congestion | Emerges from agent interaction | Assumed average speed |
| Modes | Multimodal | Freight only |
| Energy and charging | Not modelled by default | Core capability |
| Electricity network | Not represented | Represented |
Neither subsumes the other, which is why both appear in this chapter.
Role in model integration
For the purposes of model integration, selected MATRA outputs can be used to characterise aspects of transport demand, particularly:
- Freight vehicle activity
- Road-freight movements
- Fleet composition
- The spatial and temporal requirements for freight charging infrastructure
That handover is documented in freight demand.
Inputs
Spatial and operational information
MATRA requires spatial and operational information describing:
- Freight depots
- Customers
- Charging points
- Electricity substations
together with a digital road-network dataset.
Depot-level operational parameters
Used to characterise freight demand and logistics operations, including:
- Daily orders
- Delivery characteristics
Fleet and simulation parameters
These define:
| Parameter | Covers |
|---|---|
| Vehicle technology | The fleet's drivetrain mix |
| Energy consumption | How much energy movement requires |
| Battery capacity | Range before charging is needed |
| Operating costs | The economics of each option |
| Charging characteristics | Rates and constraints |
| Depot operating times | When vehicles can come and go |
| Scenario assumptions | What is being tested |
Empirical where possible, synthetic where not
Where empirical depot information is available, these data can be used directly to parameterise the model. Synthetic or hypothetical representations are generated where empirical information is unavailable.
This mirrors the approach described in freight demand, and it means results should be read with an eye to which depots were real and which were assumed.
Outputs
MATRA produces detailed vehicle-, route-, depot-, charging- and infrastructure-level outputs.
Route logs
Route logs record variables including:
| Variable |
|---|
| Vehicle identity |
| Origin and destination |
| Departure and arrival times |
| Travelled distance |
| Remaining battery state |
| Charging time |
| Queuing time |
Charging outputs
Charging outputs describe the use of individual charging locations and bays.
System-level outputs
System-level outputs report indicators such as:
- The active EV and ICE fleet
- Service level
- Unserved freight demand
- Orders per vehicle
- Charging-related queuing
- Electricity supplied at depots and public charging points
- Energy obtained from electric-road infrastructure
- Infrastructure capacity
What the outputs enable
These outputs allow alternative freight-electrification scenarios to be compared in terms of:
| Dimension | Question answered |
|---|---|
| Operational performance | Does the fleet still deliver on time? |
| Transport activity | How much movement is required? |
| Energy requirements | How much electricity, and when? |
| Infrastructure utilisation | Is the charging network sufficient? |
No public reference yet
The handbook notes that a MATRA-specific external publication or repository should be added if one is intended for public use. None is currently cited.