Forum Discussion
Identifying Broken Shifts
Hi Everyone,
I have a list of shifts and the Employees. I want to identify all the shifts that has less than 8 hours gap between them for each Employee.
sample: I want to calculate the last two column (Is broken Shift?, Broken Shift ID). does anyone can help me to deal with it?
| Start Date Time======== | End Date Time======= | Shift ID | Emp ID Adj | Is Shift broken? | Broken shift ID |
| 26/04/2021 7:30:00 AM | 26/04/2021 3:00:00 PM | 1006073 | 204492 | ||
| 27/04/2021 10:00:00 AM | 27/04/2021 1:00:00 PM | 1006169 | 204492 | Y | 1006169 |
| 27/04/2021 3:00:00 PM | 27/04/2021 9:30:00 PM | 1006074 | 204492 | Y | 1006169 |
| 28/04/2021 3:00:00 PM | 28/04/2021 10:00:00 PM | 1000915 | 204492 | ||
| 29/04/2021 3:00:00 PM | 29/04/2021 9:30:00 PM | 1006076 | 204492 | Y | 1006075 |
| 29/04/2021 6:30:00 AM | 29/04/2021 9:30:00 AM | 1006075 | 204492 | Y | 1006075 |
| 29/04/2021 9:45:00 AM | 29/04/2021 2:00:00 PM | 1008080 | 203374 |
You can find the larger sample data here!
Thanks for your help.
- Anonymous5 years ago
Just put this into the Advanced Editor in Power Query and investigate the steps one by one... This does what you want. How to extract the relevant shift id's from it? That's what I leave to you as an exercise 🙂
// Shifts let Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("pZBBCsQgDEWvIq4L/YlVa3ejxyi9/zVGKBRnkkUlu/DJ4yfvPD2DacW2cnLIR8AB+GVMKfboTglIyKFPH38tA5sd4dka0/DDUioqGzWW6bnm7t0ku6vs7pjHXhSKki0OSf7b0/LXq7Jab5E3J8lOea4Gz5J977kaPFeDZ8FOeK4Gz83gWbLvPTeD52bwLNgJz529vg==", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Start Time" = _t, #"End Time" = _t, ShiftID = _t, EmpID = _t]), #"Changed Type" = Table.TransformColumnTypes(Source,{{"Start Time", type datetime}, {"End Time", type datetime}, {"ShiftID", Int64.Type}, {"EmpID", type text}}), #"Sorted Rows" = Table.Sort(#"Changed Type",{{"EmpID", Order.Ascending}, {"Start Time", Order.Ascending}}), #"Added Index" = Table.AddIndexColumn(#"Sorted Rows", "Index", 0, 1, Int64.Type), #"Reordered Columns" = Table.ReorderColumns(#"Added Index",{"Index", "Start Time", "End Time", "ShiftID", "EmpID"}), #"Added Custom" = Table.AddColumn(#"Reordered Columns", "PrevShiftEndTime", each List.First( Table.SelectRows( #"Reordered Columns", (inner) => inner[Index] + 1 = [Index] and inner[EmpID] = [EmpID] )[End Time], null ) ), #"Added Custom1" = Table.AddColumn(#"Added Custom", "Delta < 8 hrs", each [Start Time] - [PrevShiftEndTime] < #duration(0, 8, 0, 0) ), #"Added Custom2" = Table.AddColumn(#"Added Custom1", "Shift Broken? (1)", each if not ( [#"Delta < 8 hrs"] is null ) and [#"Delta < 8 hrs"] then "Y" else null ), #"Added Custom3" = Table.AddColumn(#"Added Custom2", "Shift Broken?", each let NextRowShiftBrokenStatus = List.First( Table.SelectRows( #"Added Custom2", (inner) => inner[Index] = [Index] + 1 and inner[EmpID] = [EmpID] )[#"Shift Broken? (1)"], null ), Output = if ( NextRowShiftBrokenStatus = "Y" or [#"Shift Broken? (1)"] = "Y" ) then "Y" else null in Output ), #"Removed Columns" = Table.RemoveColumns(#"Added Custom3",{"Index", "PrevShiftEndTime", "Delta < 8 hrs", "Shift Broken? (1)"}) in #"Removed Columns"
5 Replies
- AnonymousNot applicable
Just put this into the Advanced Editor in Power Query and investigate the steps one by one... This does what you want. How to extract the relevant shift id's from it? That's what I leave to you as an exercise 🙂
// Shifts let Source = Table.FromRows(Json.Document(Binary.Decompress(Binary.FromText("pZBBCsQgDEWvIq4L/YlVa3ejxyi9/zVGKBRnkkUlu/DJ4yfvPD2DacW2cnLIR8AB+GVMKfboTglIyKFPH38tA5sd4dka0/DDUioqGzWW6bnm7t0ku6vs7pjHXhSKki0OSf7b0/LXq7Jab5E3J8lOea4Gz5J977kaPFeDZ8FOeK4Gz83gWbLvPTeD52bwLNgJz529vg==", BinaryEncoding.Base64), Compression.Deflate)), let _t = ((type nullable text) meta [Serialized.Text = true]) in type table [#"Start Time" = _t, #"End Time" = _t, ShiftID = _t, EmpID = _t]), #"Changed Type" = Table.TransformColumnTypes(Source,{{"Start Time", type datetime}, {"End Time", type datetime}, {"ShiftID", Int64.Type}, {"EmpID", type text}}), #"Sorted Rows" = Table.Sort(#"Changed Type",{{"EmpID", Order.Ascending}, {"Start Time", Order.Ascending}}), #"Added Index" = Table.AddIndexColumn(#"Sorted Rows", "Index", 0, 1, Int64.Type), #"Reordered Columns" = Table.ReorderColumns(#"Added Index",{"Index", "Start Time", "End Time", "ShiftID", "EmpID"}), #"Added Custom" = Table.AddColumn(#"Reordered Columns", "PrevShiftEndTime", each List.First( Table.SelectRows( #"Reordered Columns", (inner) => inner[Index] + 1 = [Index] and inner[EmpID] = [EmpID] )[End Time], null ) ), #"Added Custom1" = Table.AddColumn(#"Added Custom", "Delta < 8 hrs", each [Start Time] - [PrevShiftEndTime] < #duration(0, 8, 0, 0) ), #"Added Custom2" = Table.AddColumn(#"Added Custom1", "Shift Broken? (1)", each if not ( [#"Delta < 8 hrs"] is null ) and [#"Delta < 8 hrs"] then "Y" else null ), #"Added Custom3" = Table.AddColumn(#"Added Custom2", "Shift Broken?", each let NextRowShiftBrokenStatus = List.First( Table.SelectRows( #"Added Custom2", (inner) => inner[Index] = [Index] + 1 and inner[EmpID] = [EmpID] )[#"Shift Broken? (1)"], null ), Output = if ( NextRowShiftBrokenStatus = "Y" or [#"Shift Broken? (1)"] = "Y" ) then "Y" else null in Output ), #"Removed Columns" = Table.RemoveColumns(#"Added Custom3",{"Index", "PrevShiftEndTime", "Delta < 8 hrs", "Shift Broken? (1)"}) in #"Removed Columns"- SadStatueHelper II
Hi Anonymous
thanks for your quick help!
This inner join worked well. I haven't used it in M.
It seems that works well.
Cheers, - SadStatueHelper II
HIAnonymous
Thanks for your solution. I implemented what you suggested. So I can recognise the broken shift, but I still have issue in group all the related broken shift. its hard to get Min/Max index for the related shifts. I think I need a variable with a loop to find it.
I would be happy if u can help on the next step.
Cheers,
- AnonymousNot applicable
I understand you just want a column in the base table? If this is the requirement, then this should be rather easy to calculate in Power Query. I also understand that for any given individual the shifts are never overlapping, right?
- SadStatueHelper II
Hi Anonymous ,
Thanks for your reply!
I want to define these two columns (Is broken Shift?, Broken Shift ID), no matter in DAX or Power Query.
Yes, its right shifts cannot overlap for the same Employee.
Cheers,