Weird "Trend" metric calculation in Tabular KPI

Hi Experts ,
We have a tabular model in which we started designing the KPI to show Actual metric value vs its Target comparison by an Status indicator to know how good the metric has performed. In the KPI calculation window of the tabular model , there are only
place holders to calculate Value , Status & Target but not for “Trend” . Even though we didn’t code anything specific for “Trend” calculation , Under the newly created KPI we see the “Trend” along with Value , Goal & Status . But the “Trend”
is behaving weirdly in the Tabular KPI, The trend indicator is being showed for every dimension attribute that is sliced with the KPI irrespective of whether it has Metric value or not. I searched many websites to understand how this “Trend” is
being calculated in a KPI , but none of them are able to throw some light on the “Trend” calculation. In this kind of scenario please suggest me a way to circumvent this issue.
How to hide the “Trend” indicator from the newly created KPI, as this I think we cannot define “Trend” calculation in tabular as in that of Multidimensional Cubes
Understand the reason why “Trend” is displayed in tabular models
Below is snapshot of our KPI when interfaced through Excel.
Can you guys please help on how to hide the “Trend” expression from tabular models. So that our users wouldn’t be confused by using unrequired metric in KPI.
Rajesh Nedunuri.

Hi NedunuriRajesh,
According to your description, since you haven't specified any expression for Trend calculation, you want to hide the Trend option. Right?
In Analysis Services Tabular, Value, Goal Status and Trend in KPI are based on the Base Value, Target Value and Status Threshold. No matter you specify a Trend Expression or not, the Trend box is always displayed in the KPI pane. And it will do the
calculation automatically. This is feature by design. There's no way to edit or modify it. So your requirement can't be achieved currently.
I recommend you
a feature request at https://connect.microsoft.com/SQLServer so
that we can try to modify and
expand the product features based on your needs.
Best Regards,
Simon Hou
TechNet Community Support

Similar Messages

Reliability analysis metric calculation executable stopped working and was closed

JUST BROWSING WHEN RECEIVING THIS ERROR: Reliability analysis metric calculation executable stopped working and was closed
I HAVE TO COMPLETELY TURN OFF COMPUTER TO RESOLVE

JUST BROWSING WHEN RECEIVING THIS ERROR: Reliability analysis metric calculation executable stopped working and was closed
I HAVE TO COMPLETELY TURN OFF COMPUTER TO RESOLVE

Metric Calculation - Using Group By - 10g

I have the following Facts and Dimensions:
Fact1, Fact2, Fact3
Dim1, Dim2, Dim3
I want to calculate the following Metrics:
Metric1 = count(Fact1) group by Dim1, Dim2, Dim3
Metric2 = Sum(Fact2) group by Dim1, Dim2, Dim3
Metric3 = Avg(Fact3) group by Dim1, Dim2, Dim3
Metric4 = sum((Metric1/Metric2)*Metric3) group by Dim1
I have to display Metric4 in my report.
I need help to achieve this functionality
What I have tried so far:-
In RPD -
1) Create Metric 1 to 3 using evaluate and partition by clause.
----eg: Evaluate('count(%1) over (partition by %2,%3,%4)', Fact1, Dim1, Dim2, Dim3)
----Issue - Since the datatype of dimensions and fact are not similar, I am getting an error [nQSError: 22027] Union of non-compatible types.
2) Using level base measure. Setting logical level to appropriate dimension.
Issue - Unable to apply group by on Metric 4.
Any pointers on this will be appriciated.
Thanks,
Kansal

I think you are making this harder than it needs to be. You should read about the 'BY' keyword here:
http://gerardnico.com/wiki/dat/obiee/group_by#alternative_syntax_the_by_clause
Your report should just be a simple case of pulling Dim1, Dim2, Dim3, Fact1, Fact2, Fact3 in Answers and doing a calculation to get Fact4.
Paul

Use SSAS Tabular KPI in SSRS report with groups

I have a SSAS tabular model that I have built KPIs in. I have tried several attempts to use the KPI in SSRS with multiple groups, but as you probably know when you aggregate the KPI it does not calculate correctly. How do you use SSAS KPIs with SSRS groups?
Thanks in advance.

Hi Micheal,
According to your description, you want to use the KPI created in SSAS tabular for SSRS multiple groups. Right?
In Analysis Services, when we create KPI, it works for each detail record. This KPI will not consider any aggregation value. So when you aggregate the KPI in SSRS, it can't return correct result. However, in Reporting Services, you can use Variable
to define a "KPI" and apply condition with aggregation function(like sum, average) based on this "KPI" in Expression to return the correct result. Also you can use the Indicator in SSRS to achieve this requirement. Please refer to
links below:
Report and Group Variables Collections References (Report Builder and SSRS)
Indicators (Report Builder and SSRS)
If you have any question, please feel free to ask.
Best Regards,
Simon Hou
TechNet Community Support

Seasonal + Trend forecast calculation

all,
We are using Consumption based planning with MRP type VV and Forecast model - X and material master maintained as below.
For testing purpose, I have maintained past consumption for last 6 month as below screen shots.
When I execute forecast from Material master tab, I got basic values, MAD, safety stock & error total as below..
How System is calculating above values?. I have gone through the formula but didn't understood and also gone through below link but didn't get sna idea.
http://http://scn.sap.com/community/erp/post!input.jspa?containerType=14&container=2110&content Type
And How System calculate below corrected forecast values?
Please help to understand calculation.

Hello Mr,
Please check below link for seasonal trend calculation formula, Insert values and try to cross check result.
I hope this link helpful to you.
Forecast with Seasonal Trend Model - Forecasting - SAP Library
https://help.sap.com/saphelp_scm70/helpdata/en/46/a605b9d57052b8e10000000a155369/content.htm
Thanks & Regards
Umesh Mali

Trend formula calculation in Calc

In the MS Excel I was using before it was easy to create a simple scattered chart, and then to select a trend option in order to draw a trendline of a given type along with its formula. I really need this functionality in calc, but do not know how-to.
Thank you,
Mike

Hello Sachin,
Please check the possibilities of the IMG activity 'Specifiy Events'.
Example: event EVT_EXPL06 (Explosion: List of Ingredients (Specifications)) for the Ingredients view and component Z_PACK for packaging. With the indicator Explosion and Calculations it is possible to control the usage.
Please have a look into the documation of the IMG activity.
Best regards,
Roland

Metrics Calculation in Scom 2012

Hi,
I have a question to ask. Please help
When ever SCOM monitors this infrastructure metrics (like CPU, Memory,Disk,Network) by deploying Custom MP's or default   ones available, how the calculations are done to get the values.
   If i would like to know how my CPU utlization is calculated, how can i know?
   Is there any doc or forum available where i can get this info? Please help.
ShivaKumar.G

Hi
SCOM usually uses the Perfromance Counters which come with Windows or other products, SQL etc. check this here for brief introduction
http://technet.microsoft.com/en-us/library/hh457548.aspx . Of course you could build a collection rule based on a script and collecting data. But this is a different story
http://technet.microsoft.com/en-us/library/ff629410.aspx.
Cheers,
Stefan
Blog: http://blog.scomfaq.ch

Opportunity and metrics calculations used in reports

Hi,
Need to know how various entities are being calculated in the Opportunity and revenue matrices.
Eg:- what's the formula being used to calculate Avg. Sales Cycle and Avg. Days in Stage etc.
Any suggestions will be helpful.
Thanks
Nakul

Hi Nakul,
I know that for the Sales Stage report that each night when the ETL runs it takes a snap shot of the opportunities and what sales stage they are in (etc), and the Avg. Days in Stage is based off all the snapshots for how many days a opportunity was in a sales stage.
I’m sorry though I don’t know what the formula’s are that are actually used to figure out 2 two metrics you asked about.
Hope this helps.
-Joe

[ASO Calculation] Weird behavior of calculation in ASO

Hi,
I have a strange scenario in one of ASO cube (in 11.1.2.3 ) where calculation script (highlighted in orange below) written and run with POV set for all time periods results in expected values only for one period and for the rest results are not seen (we see only #MISSING). We have similar logic for other calculation and it seems to have worked fine. I am puzzled now as to why the calculation (with similar logic) work for all periods in one script and doesnt in other script. What is preventing results to be seen across all periods ? We have tried
(a) slice merge - NO CHANGE, (b) clear partial data > re-run calc > slice merge - NO CHANGE. We have run out of options now and have no clue as to what is triggering this strange behavior in ASO ? Is this a bug ?
=========================================================================================================================
Here even-though we have POV set for all periods, the result is only seen on Oct-15 and not for other periods.When we do manual calculation in Excel, we see result but from the cube we get only #Missing.
CALCULATION RESULTING IN PARTIAL RESULT:
[L0_S/S COS per Day] := (([AC_ZZ5030],[DP_Z3100]) + ([AC_ZZ5020],[DP_Z3100]) - ([AC_500510],[DP_Z3100]) - ([AC_500520],[DP_Z3100])) / ([AC_991023],[DP_Z3100]);
MAXL RESULTING IN PARTIAL RESULT:
execute calculation on database "WAG_FBI"."WAG_FBI" with
local script_file "/u01/app/test/fmw/instances/instance1/Essbase/essbaseserver1/app/WAG_FBI/WAG_FBI/SS_DOS.csc"
POV "Crossjoin({[Loaded]},
Crossjoin({[Actual]},
Crossjoin ([Time].Levels(0).Members,
Crossjoin({[DP_15125]},
Crossjoin([LOB].Levels(0).Members,
Crossjoin([LE].Levels(0).Members,
[Location].Levels(0).Members))))))"
SourceRegion "CrossJoin({[AC_ZZ5030],[AC_ZZ5020],[AC_500510],[AC_500520],[AC_991023]},{[DP_Z3100]})" ;
=========================================================================================================================
Here we have same POV and similar calculation logic (only change being in the denominator and the account where result should be stored) , the result is seen across all periods.
CALCULATION WHERE COMPLETE RESULT IS SEEN:
[L0_Rx COS per Day] := (([AC_ZZ5030],[DP_10010]) + ([AC_ZZ5020],[DP_10010]) - ([AC_500510],[DP_10010]) - ([AC_500520],[DP_10010])) / ([AC_991012],[DP_10010]);
MAXL WHERE COMPLETE RESULT IS SEEN:
execute calculation on database "WAG_FBI"."WAG_FBI" with
local script_file "/u01/app/test/fmw/instances/instance1/Essbase/essbaseserver1/app/WAG_FBI/WAG_FBI/RX_DOS.csc"
POV "Crossjoin({[Loaded]},
Crossjoin({[Actual]},
Crossjoin ([Time].Levels(0).Members,
Crossjoin({[DP_10010]},
Crossjoin([LOB].Levels(0).Members,
Crossjoin([LE].Levels(0).Members,
[Location].Levels(0).Members))))))"
SourceRegion "CrossJoin({[AC_ZZ5030],[AC_ZZ5020],[AC_500510],[AC_500520],[AC_991012]},{[DP_10010]})" ;
=========================================================================================================================
Regards,
Sathish

I have no idea what the cause of your problem is, but a suggestion for a test...
Create a member (e.g. 'Calc L0_S/S COS per Day') in the same dimension as [L0_S/S COS per Day] (assuming this a dynamic dimension), assign the formula to it (instead of placing it in the script) then make your script copy from that member to the target.
I felt sure I had seen a defect that related to partial ASO allocation results but I can't find it now, so perhaps I was confused.

Oracle 10g - Metric Calculations

I am trying to support a query that creates a report(hourly) to be used in Excel and some of the stats may be incorrect.A user is prompted to enter a start date from when to begin searching for records.The goal is to generate measurements on a process. The process begins when a customer has added his email to our website. Then he is sent an email for verification.However,problems sending the email may occur indicated by the TRANS_STATUS codes. DATE_EMAIL_ADDED is when the customer fills out our website and submits his email. DATE_EMAIL_SENT is when the email was validated and the time the notification was sent. The dates are timestamps. Also, I would like some thoughts on the calculations to make sure they are correct. Any suggestions on optimizing this query or a different approach to the solution is appreciated.
-- DDL for Table TRANS_DTL
{code
CREATE TABLE "TRANS_DTL"
(     "TRANS_ID" NUMBER(22,0),
     "TRANS_TYPE" VARCHAR2(5),
     "START_TIME" DATE,
     "END_TIME" DATE,
     "TRANS_STATUS" VARCHAR2(3)
code}
-- DDL for Table EMAILINGS
{code
CREATE TABLE "EMAILINGS"
("ATTEMPTS_MADE" NUMBER,
     "DATE_EMAIL_SENT" DATE,
     "DATE_EMAIL_ADDED" DATE
select
A.DATE_EMAIL_ADDED,
A.NUM_EMAIL,
A.EMAIL_TYPE,
A.AVG_TIME_TO_EMAIL,
A.FIRST_ATTEMPT_RATIO,
B.TOT_TXNS,
B.TXN_Success_Rate,
B.TXN_API_Exception_Rate,
B.TXN_System_Exception_Rate,
B.AVG_Duration
FROM
(select TO_DATE(to_char(DATE_EMAIL_ADDED,'MM/DD/YYYY HH24'),'MM/DD/YYYY HH24') DATE_EMAIL_ADDED,
w.email_type EMAIL_TYPE,
count(*) NUM_EMAIL,
avg((DATE_EMAIL_SENT - DATE_EMAIL_ADDED) * 1440) AVG_TIME_TO_EMAIL,
sum(case when ATTEMPTS_MADE = 1 then 1 else 0 end) / count(*) first_attempt_ratio
from EMAILINGS W
where DATE_EMAIL_ADDED >= to_date('&date','MM/DD/YYYY')
group by TO_DATE(to_char(DATE_EMAIL_ADDED,'MM/DD/YYYY HH24'),'MM/DD/YYYY HH24')
,w.email_type
) A,
(select
Start_Date,
count(*) TOT_TXNS,
sum(case when TRANS_STATUS = '000' then 1 else 0 end) / count(*) TXN_Success_Rate,
sum(case when TRANS_STATUS = '997' then 1 else 0 end) / count(*) TXN_API_Exception_Rate,
sum(case when TRANS_STATUS = '999' then 1 else 0 end) / count(*) TXN_System_Exception_Rate,
avg(elapsed_seconds) AVG_Duration
from (
select
to_date(to_char(Start_Time, 'MM/DD/YYYY HH24'),'MM/DD/YYYY HH24') Start_Date,
TRANS_STATUS,
trunc(to_number(substr((a.End_Time-a.Start_Time),1,instr(a.End_Time-a.Start_Time,' ')))) *86400 +
substr((a.End_Time-a.Start_Time),instr((a.End_Time-a.Start_Time),' ')+1,2)*3600 +
substr((a.End_Time-a.Start_Time),instr((a.End_Time-a.Start_Time),' ')+4,2)*60 +
substr((a.End_Time-a.Start_Time),instr((a.End_Time-a.Start_Time),' ')+7,2) +
substr((a.End_Time-a.Start_Time),instr((a.End_Time-a.Start_Time),' ')+10,6)*0.000001 elapsed_seconds
from TRANS_DTL a
where
TRANS_TYPE = 'ESP'
and Start_Time >= to_date('&date','MM/DD/YYYY')
and end_time is not NULL
group by Start_Date) B
where A.DATE_EMAIL_ADDED = B.START_DATE
order by B.START_DATE;
DATE_EMAIL_ADDED NUM_EMAIL EMAIL_TYPE AVG_TIME_TO_EMAIL FIRST_ATTEMPT_RATIO TOT_TXNS TXN_SUCCESS_RATE TXN_API_EXCEPTION_RATE TXN_SYSTEM_EXCEPTION_RATE AVG_DURATION(sec)
1/10/2011          116               WE          5.59               99.14               127               89.76%          10.24%               0.00%                    9.31
code}

Nicosa wrote:
No, it uses curly brackets, like that {noformat}{noformat}.No, it can be both. either {noformat}{noformat} before and after the code OR {noformat}[code]{noformat} before and {noformat}[/code]{noformat} after.

Drill Down in Tabular Format Metrics

I know how to drill down inside of a Pie Chart for Metrics using the drillDown() method and passing the arguments.
But how do I do this, when my Metrics are in Tabular format displayed as a Group on the dashboard to the end user..
So when the end-user clicks on one of the rows lets say employee ID, it should pop up another tabular format Group presentation show the address, department etc.. of that employee?
How do I do that?
I don't have any On-Click Method where I can specify the drillDown() method for my tabular Format Group Metrics?

The "on change invoke" property's method is able to automatically detect the row in your group that was clicked because your method is inside your group and not in the parent BPM Object. If you're on 10g, make sure you've upgraded to a recent hotfix for 10g, there is a bug that prevents this from working on earlier cuts of the software.
Inside your group, add a new method called "showMeTheNameIChanged()". Add logic similar to this:
// you do not need this unless you want to determine the row # that was clicked
//   - this assumes that your group's name is "orderLines"
lineSelected = orderLines.indexOf(this)
// this will display the value of the attribute "customerName" in the row changed
display "You changed the name on the row ["
        + lineSelected + "] (zero based) and changed the name to ["
        + customerName + "]"Because the method is associated with the group and not the parent BPM Object, it automatically knows that you want the attributes for the row that was clicked. Again, although you can retrieve the row number ("lineSelected" in the above logic) you do not need to do this to retrieve the values of the attributes in the row changed ("customerName" in the above example).
Dan

Large tabular cube w/ lots of measures -- keyboard stuck

hi folks,
I was wondering if anyone has seen this weird behavior: on fairly large tabular cubes (SSAS 2012 SP2), with lots of calculations, the keyboard seems to be stuck, one cannot type or make any modifications to any measures, even after closing a project.
The only workaround is to completely restart the machine. Using Visual Studio 2010 Shell.
Are there any possibly corruptions or calc limitations that one should be aware of?
Notable that this never happens with any small cubes, but just with the largest cube that we have, that has probably tens of calcs.
any thoughts are much appreciated,
Cos

Hi Cos2008,
According to your description, your keyboard get stuck when processing a huge tabular. However, it works properly when processing a small tabular. Right?
In this scenario, I think the keyboard stuck you mentioned should be the entire machine stuck. Since this only happens for big tabular, I guess it should be caused by lack of hardware resource. Please reserve enough disk space for the .cub file. And
set the memory configuration for SSAS engine and SQL instance:
Server Properties (Memory Page)
Memory Properties
Also please refer to a white paper below to optimize your SSAS tabular.
Performance Tuning of Tabular Models in SQL Server 2012 Analysis Services
Best Regards,
Simon Hou
TechNet Community Support

Reg:SSM KPI Drill Down (Once Click KPI need to display relevant measures)

Dear Friends
I have the enclosed requirement. Pl. suggest proper solution.
When we drill down need to display the calculated measures of respective KPI.
Here same KPI has different measures in drill down.
Does we need to have different KPIs or can we have different measures for single KPI as per dimension member
Pl. suggest
Rgds
SriG

Sri,
Are you saying that you have a KPI that has 2 dimensions but the 2 dimensions have completely different metric calculations so that the component metrics that that are combined to make your DIMM1 have no influence on you DIMM2?
If this is the case then I think you would need 2 separate KPIs which would have different structures.
There is no problem combining multiple metrics together to create a KPI and even incorporating calculations into that combination.
To use an example from the insurance world: Combined Operating Ratio is derived by combining net claims, net commissions and net expenses, each of which is derived from multiple data elements and each of these metrics can be dimensioned by for example product, channel, region.
Regards
Colin

S&OP factor calculation (alpha, beta, gamma)

Hi gurus,
I´m trying to adap the automatic forecast selection in MC88 transaction to a graphic company scenario. They have several clients where they planned all the forecast for them, based on increase/decrease sales of each client. Today they are using the alpha, beta, gama factors but they still need to calculate the demand in a Excel spreadsheet and adjust the result in SAP.
My questions are:
- There´s any way to change the factors criteria? For example, if I want to increase the forecast for 2012 in 10% over 2011. If not, anyone knows a way to input this growing factor into the automatic forecast?
- I already found the formulas for the alpha, beta, gamma, but it´s not didactic to explain to some users. Anyone knows where I can find a simple and objective explanation for this factors?
Thanks in advance.
Best Regards,
Felipe Tavares

Hi,
It's your wrong assumption that clients are not using Forecast. Most of the Retails clients are effectively using forecast.
Without forecasting, Retail operation will not work at all.
A demand forecast is a prediction of the amount of each article / product that will be sold in the future. It's easy to see that inaccurate forecasts can cause major problems for your company. When forecasted demand is far below actual demand for a product, you risk running out of products and disappointing your customers. If forecasted demand exceeds actual usage, you'll probably be faced with large quantities of dead stock and slow-moving inventory.
For your company to be successful, your demand forecasts for products must be as accurate as possible.
(a) Alpha is used in the basic value calculation
(b) Beta is used in trend value calculation
(c) Gamma is used in the Seasonal index calculation
The value for the parameters range from 0 to 1. A higher value will place more emphasis on recent history. The parameters also control how reactive the forecast is to changes in historical patterns.
This is FYI.
Bye,
Muralidhara

Attention All Metric Halo and ChannelStrip Users - What's happening?

Hi Everybody ...
About 2 months ago I posted that ChannelStrip for GarageBand was an amazing addition to the capabilities of GarageBand ...
I had tried the demo and was very impressed ....
Well ... just last week I went online and bought ChannelStrip for GarageBand...
but since then something very weird with Metric Halo is going on ..
I paid my $89 for ChannelStrip ... and I haven't heard a word from them since.
No email confirmation .... nothing ...
I sent 2 emails to them asking where my purchase was over their web support page ... no response ...
Try calling their support line ...http://www.mhlabs.com/metric_halo/support/.. all you get is a bunch of extensions and weird operator voice mail ... I left various voice mails ... and again ... no response ...
I'm starting to think that this company has gone under ... but they are still receiving money over the internet for nonexistent services ...
Has anyone heard anything about Metric Halo ????

I just got an email a couple of days ago and last week received email support within a day. I also use ChannelStrip.
I also called the same support number and got through on the first ring two weeks ago.
The phone line message is normal, as long as I've been a customer.
But I would say your experience of support isn't normal. They have always been known for their quality support.
In fact when I was looking for my ULN-2, they went out of their way to get one to me quickly.
Perhaps post here in the hardware section:
http://www.bigbluelounge.com/forums/index.php
There are a lot users there who may have a better number or better info. I'm sorry but I've misplaced the direct number I had.
I've dealt with Jason Desmond and Jon Stern at MH support. And like I said, they have always been great to deal with and seem to go out of their way to assist. Not to mention how great their products are.
Perhaps the holidays this week and last have had an impact....

Weird "Trend" metric calculation in Tabular KPI

Similar Messages

Maybe you are looking for