Читать книгу People Flow in Buildings - Marja-Liisa Siikonen - Страница 25

Manual counting gives random samples of the daily passenger traffic in buildings. Counting can be performed at elevator lobbies, inside elevators, or for instance from closed‐circuit television (CCTV) video recordings.

Pen and paper method does not require special technological knowledge or investment in technical devices. To find out vertical traffic, checkers can register the passengers getting on and off the elevators with time stamps, see Table 2.1. Additional information such as if the car was travelling up or down may be added. The data is later analyzed for desired intervals, which is called the post‐sampling method. This method is flexible for research purposes. Another alternative is the pre‐sampling method, where passenger arrivals are counted for predefined intervals.

Table 2.1 Minimum information for counting passengers in an elevator lobby.

Building information, lobby, floor index, checker
Time stamp	Enter	Exit	Group size
8:01	2	0	1
8:03	1	1	4
8:04	3	0	1
…

If there are several entrance floors, counting should be done simultaneously at each entrance. From this data, the floor or building occupancy during the measurement period can be estimated. Manual counting at the main entrances does not include the traffic on upper floors nor the information of passenger journeys. To be able to count the inter‐floor traffic on upper floors, persons should stand in the lobbies of each level and record the passenger transfers in to and out from the elevators with time stamps. The direction of the car and floor index should be included in the data. A less laborious approach is if there is a traffic checker in each car who writes down the passenger transfers in to and out of the car with time stamps during each stop. Also, the floor index as well as the direction of the car should be recorded. For passenger journey information, the origin and destination floors of each passenger should be recorded as well, but this is quite a demanding task for one person.

The problem with manual counting is the accuracy, consistency and reliability of the data with several checkers. Capable data collection resources are often limited since the counting is labour‐intensive and time‐consuming. From the random samples of daily traffic, it is difficult to form trends or forecasts of the traffic.