Hi KNIME Community ,
I would like to delete the resting time window of the machine such as 10:00AM to 10:15 AM and 12:00 to 13:00 and also Saturday and Sunday from the time series data.
Please also note that the rest time may not exactly start at 10:00 AM or 12:00 AM etc., but ±6 min earlier or later. I would appreciate your help.
1 Like
Hi Marzukim,
Thanks for your time. I would appreciate a little explanation on each node functionality
Regards,
Hi Marzukim,
Thanks for your time. As I mentioned in the question, the machine Stop/Start at lunch break few minutes earlier or later than 12:00 PM. For example, the machine shutdown some days 11:53AM or earlier and start at 13:09 or later. So, I need to identify those time period and delete from the data set.
Thank you in advance
Hi @Vaamen and welcome to the KNIME community.
If you are able to upload a small sample of your data, such as in an excel file, (with any sensitive information removed), that will help people to assist you, as it not only makes it easier for people to visualise your data and avoid wrong assumptions but also gives people something to try out their workflow ideas on. thanks.
Hi @takbb ,
below is the sample , need to delete the lunch break time from 1 year data set
Start Time / End Time/ Time Diff
11:36:42 AM 11:38:54 AM 2
11:42:12 AM 11:43:16 AM 1
11:45:14 AM 11:45:28 AM 0
11:50:40 AM 13:06:26 PM 75 (Lunch break)
13:17:46 PM 13:20:50 PM 3
… . …
… …
Hope you get the picture
thanks
1 Like
hi @Vaamen
i’m not an expert at explaining things, especially in english. 
referring to my previous post since don’t have any real data, i just created dummy data to illustrate how the filter works. extracted and split the data using three nodes before the filter node. the required nodes depend on how your raw data is presented.
to answer to your recent question:
- if your data pattern for machine shutdown during lunch break always occurs between 11:50 am and 13:10 pm, you can set rules to:
$YourDate&TimeColumns$ >= “11:50” AND $YourDate&TimeColumns$ <= “13:10” => TRUE - b. but if your data already states the “lunch break” in the description, it becomes easier: you can just filter through patterns.
${YourDate&TimeDescription} LIKE “lunch break” => TRUE - c. to filter out days such as sunday, saturday, etc., you must mention either the date or the name of the days in the raw data.
I’ve attached a sample workflow for you to explore and start with.
KNIME_time-windw-MM.knwf (142.6 KB)
If you’re using version 5.1, you can get a detailed explanation of any node by clicking the description tab on the left of the panel.
If you need more information about KNIME nodes, this forum link may be a helpful starting point:
rgds
linux knime 5.1.x
2 Likes
Hi @marzukim ,
Thank you very much for working around my queries. I appreciate your suggestions.
However, the hold here is the break time not always start at same time, some time little early or late.
In addition, there is no “lunch Break “entity in the table
thanks
it doesn’t matter whether your lunch break happens early or late, as long as it’s within the boundaries of your given ‘rules’ time. systems doesn’t know when your lunch break is, so you need to specify a condition or statement.
or another option that you can use, is to use the following syntax:
$S-am/pm$ =“AM” AND $E-am/pm$ = “PM” => TRUE
this will consider any data that contains AM switching to PM to be lunch break filtered data.
your lunch break cannot be other than noon time.
happy knimining
rgds
linux knime 5.1.x