Accommodate Last-Minute Schedule changes

Product Used: Route Optimization API

Background

An NEMT provider operates in a large urban area, transporting patients to various medical facilities for treatments and routine doctor visits. Every day, the provider handles a large number of scheduled trips, with multiple pickups and drop-offs throughout the day.

On a typical morning, all vehicles are dispatched based on an optimized route plan generated at the beginning of the day. However, as the day progresses some unexpected changes occur:

Last minute appointment changes: One of the medical facilities extends the appointment duration for a patient just 5 minutes before his scheduled pickup. This creates idle time for the assigned driver, leading to inefficiency.
New urgent requests: A dialysis clinic contacts the provider with an urgent request to transport a patient who needs to get to an appointment within the next hour. However, the nearest available driver isn’t initially scheduled to handle this new request.
Unexpected delays or no-shows: A driver arrives at a patient’s home, but the patient is not ready and needs more time. The driver cannot wait indefinitely because other patients also need to be picked up.

These last-minute changes can throw off the entire day's carefully planned routes, leading to delays, inefficiencies, and the potential for missed appointments for other patients.

Problem Statements

The NEMT wants to efficiently manage:

Last minute appointment changes: As the treatment center extends the scheduled duration of the appointment, drivers are left with gaps in their schedule, wasting time and resources.
New, Last-Minute Requests: Medical facilities or patients often make last-minute requests, requiring immediate transportation, but the current route plan does not account for these new demands.
Need for Route Re-Optimization: As these changes accumulate, the initial route plan becomes outdated, and drivers are no longer operating on the most efficient or optimized paths, leading to longer wait times, increased costs, and potentially missed appointments.

Solution

NextBillion.ai Route Optimization API allows the NEMT provider to re-optimize routes in real time when changes occur. This dynamic re-optimization ensures that the fleet operates efficiently even when faced with last-minute schedule changes.

Last minute appointment changes

We add the entire existing route plan into the solution attribute of a new input request. While configuring the pickups in the new request we remove the pickup for the patients whose appointment duration has been extended.

The optimizer will tweak the existing route plan minimally to reflect the new changes

New last minute requests

When re-optimizing, we add the new patient’s pickup and drop-offs into the input problem request along with its `priority` settings. The optimizer assigns this new request to an existing vehicle based on its proximity to the new location and other constraints to generate a new, efficient route plan.

Need for Route Re-optimization

Routes can be re-routed dynamically as more and more such scenarios occur throughout the day. When re-optimizing for trips already in-progress, just remove the tasks that are already completed from the new inputs and update the vehicle's capacity to reflect the current capacity at the time of making the request.

The Directions API considers real-time traffic and historical traffic to compute the best routes such that unnecessary detours and delays due to traffic congestion are avoided. Consequently, efficient routing helps the provider minimize fuel consumption and operational costs.

Sample API request and solution

Here is complete sample JSON request that we built for this case and the corresponding response:

Original Solution

We have an original solution which uses 2 vans to pick up several patients from a treatment center and drop them off at their respective homes. The vans are at the treatment center and waiting for all scheduled patients to come out of their sessions. In a short time following changes happen

A patient whose session was supposed to be over in the next 5 mins, will need to wait another hour due to some new complications during the session.
There are another set of new patients, who just got to the treatment center, had their sessions canceled and now need to be taken back to their homes. These are treated as new urgent requests for drop-offs.

So we add new patients (`id`: Patient Drop-off 5, Patient Drop-off 6, Patient Drop-off 7) and remove an existing patient (`id`: Patient Drop-off 1) and re-run the optimization for updated route plans.

Re-optimization API request

1
curl --location 'https://api.nextbillion.io/optimization/v2?key=<your_api_key>' --header 'Content-Type: application/json' --data '{...}'

Re-optimized Solution

1
{
2
    "description": "Accommodate Last Minute Changes",  
3
    "result": "{...}",
4
    "status": "Ok",  
5
    "message": ""  
6
}

We can see in the re-optimized solution, that the patient 1’s drop-off has been removed while the drop-offs for patients 5, 6, 7 were added to the final route plan. The re-optimization feature of the Route Optimization API allows the NEMT provider to dynamically adjust to the day’s inevitable schedule changes, ensuring continuity of service and patient satisfaction.