KPI data on a corridor into a specific time range

How to get a specific day and hour for the entire network (snapshot in the past)

The use case is focused on to get the measured speed and flow data for the entire network at a specific date and time in the past. You can understand what happened across the network at that point, aiding in event analysis and anomaly detection.

First step

The endpoint is:

GET /time-slice/get

Path parameter

Not applicable.

Query parameters

For additional details, see API Reference.

Parameter	Required	Note	Example
fromTime	YES	Start of the time interval.	2025-01-02T08:00:00Z
toTime	NO	End of the time interval. It must not exceed 1 hour after fromTime. If not specified, only data associated to fromTime is returned.	2025-01-02T08:10:00Z
fromRow	NO	Index of the first element to return (starts from 1).	1
toRow	NO	Index (not included) of the last element to return.	10

fromRow and toRow can be set to paginate large datasets.

Example of request

GET https://api.ptvgroup.tech/hda/v1/time-slice/get?fromTime=2025-01-02T08:00:00Z&toTime=2025-01-02T08:10:00Z&fromRow=1&toRow=10 HTTP/1.1

Host: api.ptvgroup.tech

Authorization: apiKey YOUR_API_KEY

Accept: application/json

Example of response

{
    "metadata": {
        "id": "-1996853140",
        "mapVersion": "20241112133222",
        "createdOn": "2025-01-15T17:50:16.982182087Z",
        "totalElements": 13214,
        "totalSize": 3118372,
        "maxElementsPerRequest": 212765
    },
    "results": [
        {
            "index": 1,
            "streetCode": "1029074164319670",
            "streetIdno": 239600,
            "streetFromNode": 198070,
            "streetToNode": 16825,
            "openLrCode": "Cwi0eR203iOOAAAN/+UjHg==",
            "mapVersion": "20241112133222",
            "values": [
                {
                    "speed": 33.0,
                    "fdat": "2025-01-02T08:06:00Z",
                    "ldat": "2025-01-02T08:06:47Z"
                },
                {
                    "speed": 33.0,
                    "fdat": "2025-01-02T08:00:00Z",
                    "ldat": "2025-01-02T08:00:46Z"
                }
            ]
        },
        {
            "index": 2,
    // Additional street results...
  ]
}

Second step

Based on the response, you can analyse the data snapshot.

In the request is specified that the response is made-up by 10 rows. In this case it's easy to recognize in the response, into the structure results, a specific sub-set of information associated with a specific index value.

In the example of the response is shown only the data associated to the index=1.

Due to potentially large data volumes:

Consider to process the data in batches, using fromRow and toRow parameters to paginate the data.
Be careful of network bandwidth and memory usage when requesting large volumes of data.
Implement efficient data processing pipelines and consider storing data in scalable storage solutions.

Best practices

About time constraints:

Ensure that toTime parameter is within 1 hour of fromTime.
If you need data for a longer period, make multiple requests in 1-hour intervals.

Data formats available are:

JSON for easy integration and parsing, especially for smaller datasets.
Apache Parquet suitable for big data processing. Use tools like Apache Spark, Hadoop, or Pandas in Python to process Parquet files.

Use Case