Operational impacts
Codes OP1 to OP6. The second link in the chain: what reduced network availability does to journeys.
Operational impacts split into two kinds. Some journeys still happen but take longer. Others do not happen at all.
Delay: OP1 to OP4
Either speed reduction or rerouting results in journey delay, which is a measure of operational performance loss.
Its significance can be evaluated using total delay, average delay, or peak-time average delay.
| Code | Metric | Applies to |
|---|---|---|
| OP1 | Total delay (min) | Short and long-term |
| OP2 | Average delay (min) | Short and long-term |
| OP3 | Morning-peak (7 to 9 AM) average delay (min) | Short and long-term |
| OP4 | Afternoon-peak (4 to 6 PM) average delay (min) | Short and long-term |
Why total, average and peak are all reported
Each conceals something the others reveal.
Total delay captures the scale of the loss across the whole system, but a large total could mean severe delay for a few or mild delay for many.
Average delay captures typical experience, but averages hide the tail.
Peak-window averages isolate the periods when the most people are travelling and when delay has the greatest knock-on effect on work, school and appointments. A disruption that lands at 08:00 is a different event from the same disruption at 14:00.
Cancellation: OP5 and OP6
When feasible travel routes become unavailable, transport users are forced to cancel their journeys, and bus services may also be suspended.
| Code | Metric | Applies to |
|---|---|---|
| OP5 | Number of cancelled journeys by trip purpose | Short and long-term |
| OP6 | Number or percentage of cancelled bus service | Short and long-term |
By trip purpose, not just by count
OP5 is disaggregated by trip purpose deliberately. A cancelled leisure trip and a cancelled hospital appointment are not equivalent losses, and they carry different economic values in EC2.
The distinction also matters socially: which purposes get cancelled tells you something about who is absorbing the disruption.
OP6 covers public transport supply rather than individual choice. A suspended bus service removes the option entirely, which particularly affects travellers without access to a car.
Required input data
Operational performance metrics are obtained directly from agent-based model outputs, for example by comparing agents' daily trips and network conditions against a baseline scenario.
Relevant agent-based transport outputs can include:
- Modal shares
- Vehicle counts
- Trip times
- Trip distances
- The spatial and temporal distribution of traffic
(Alvarez Castro, 2024)
Everything here is a comparison
Delay only exists relative to something. These metrics are the difference between a disrupted run and a baseline run of the same scenario, which is why the transport model must be executed twice.
This is the same do-nothing baseline principle described in adaptation measures.
Where this leads
Operational loss propagates in three directions at once:
| Consequence | Dimension |
|---|---|
| Rerouting and congestion raise emissions | Environmental |
| Delay and cancellation fall unevenly across groups | Social |
| Lost time and lost trips carry a monetary value | Economic |