I have a problem with calculating average time. This is the case: i have multiple rows, in each row there is data in time format so I need to calculate an average time of all rows and multiply it with number of rows but of course I have a problem with data format because I need to multiply time with integer
Can somebody help me with some advice? thnx Here is some data:
times
00:00:00.7400000
00:00:01.1870000
00:00:00.6430000
00:00:00.6100000
00:00:12.9570000
00:00:01.1000000
00:00:00.7400000
00:00:00.5300000
00:00:00.6330000
00:00:01.6000000
00:00:02.6200000
00:00:01.0300000
00:00:01.9630000
00:00:00.9800000
00:00:01.0170000
00:00:00.7600000
00:00:00.7130000
00:00:00.9730000
00:00:01.0000000
00:00:01.0530000
00:00:02.9400000
00:00:00.8200000
00:00:00.8400000
00:00:01.1800000
00:01:25.8230000
00:00:01.0000000
00:00:00.9700000
00:00:01.2930000
00:00:01.3270000
00:00:13.5570000
00:00:19.3170000
00:00:58.2730000
00:00:01.6870000
00:00:18.7570000
00:00:42.8570000
00:01:12.3770000
00:00:01.2170000
00:00:09.9470000
00:00:01.4730000
00:00:00.9030000
00:00:01.0070000
00:00:01.1100000
00:00:01.6270000
00:00:05.0570000
00:00:00.6570000
00:00:00.7900000
00:00:03.2930000
00:00:00.8600000
00:00:01.0330000
00:00:00.9300000
00:00:00.8730000
00:00:00.9600000
00:00:00.8070000
NULL
so from this data a need average time or/and sum of that data
To calculate this average, you can use AVG command. – Averages per minute: to obtain the averages per minute, you must retrieve E3TimeStamp's seconds and milliseconds. To do so, multiply this field by 24 (to convert the time base into hours), and then by 60 (to convert it into minutes).
Drawing a line between each value in the sample (including the interpolated values) Calculating the area below the line. Dividing the result of the area calculation by the duration of the sample. The result of this step is the time average for the sample.
AVG () computes the average of a set of values by dividing the sum of those values by the count of nonnull values. If the sum exceeds the maximum value for the data type of the return value, AVG() will return an error.
Answer: To find the average time between two dates, you could try the following: SELECT TO_DATE(date1, 'yyyy/mm/dd') + ((TO_DATE(date2, 'yyyy/mm/dd') - TO_DATE(date1, 'yyyy/mm/dd')) /2 ) FROM dual; This will calculate the elapsed time between date1 and date2.
You should be able to use something similar to the following:
select
cast(cast(avg(cast(CAST(times as datetime) as float)) as datetime) as time) AvgTime,
cast(cast(sum(cast(CAST(times as datetime) as float)) as datetime) as time) TotalTime
from yourtable;
See SQL Fiddle with Demo.
This converts the times
data to a datetime
, then a float
so you can get the avg
/sum
, then you convert the value back to a datetime
and finally a time
You can do this by doing a bunch of date arithmetic:
DECLARE @t TABLE(x TIME);
INSERT @t VALUES('00:01:30'),('00:02:25'),('00:03:25');
SELECT CONVERT(TIME, DATEADD(SECOND, AVG(DATEDIFF(SECOND, 0, x)), 0)) FROM @t;
However, what you should be doing is not using TIME
to store an interval/duration. TIME
is meant to store a point in time (and breaks down if you need to store more than 24 hours, for example). Instead you should store that start and end times for an event. You can always calculation the duration (and the average duration) if you have the two end points.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With