Introduction to Timeseries Data

Overall, these tutorials will provide you with the knowledge and skills necessary to retrieve and analyze time series data from a variety of assets, allowing you to gain insights into various aspects of your environment and optimize your operations accordingly.

Windowing / Function Recommendations

The recommendation table provides guidance on the appropriate window sizes and functions to use for different measurements related to occupancy monitoring. Here is a summary of the recommendations:

1. floor_occupancy | room_occupancy | zone_occupancy: For tracking (floor | room | zone) occupancy, it is recommended to use a window size of 5 minutes with either the max or mean function. This allows for capturing the highest or average occupancy level within each 5-minute interval.

   • 5 minutes smoothes out noise in the data and reduces the volume of data
   • Aggregates detections from all sensors within a floor | room | zone set-up in Studio
   • The window you choose depends on the question you are trying to answer with this data. As a general rule:
     • Max is recommended for smaller time windows at 5 minutes or less.
     • Mean is recommended for larger time windows greater than 5 minutes.

2. traffic: For measuring the flow of individuals entering and exiting a space, there is no specific recommendation for the window size. However, the sum function is recommended to calculate the total traffic, providing an aggregate count of the number of people passing through over the specified time period.

   • Sum to total entrances and exits from the start, to the end of the day.
   • Use cases:
     • Traffic for the day
       • window = time period of interest (ie. 9am-12pm)
     • Estimate occupants based on in and out counts:
       • start window from start of the day to present time.

3. presence: To monitor the presence or absence of individuals within a room or area, a window size of 30 seconds is suggested with the max function. This captures the highest presence level within each 30-second interval, allowing for real-time monitoring of occupancy changes.

   • near real-time detections for individual sensors.
   • Use cases:
     • testing purposes
     • if you want a single sensor data to feed into your own occupancy algorithm
   • Window: 30 seconds - reduces noise from raw data and captures events where a person is moving quickly across the sensor.

By following these recommendations, organizations can effectively track and analyze occupancy trends and patterns, enabling them to make informed decisions about space management, resource allocation, and other related considerations. It is important to adapt these recommendations to specific needs and adjust the window parameters accordingly.

Measurement	Window Size	Function
room_occupancy	5m	max, mean
floor_occupancy	5m	max, mean
zone_occupancy	5m	max, mean
traffic	*any	sum
presence	30s	max