Home

Key Functionalities of ERPJewels

What is new in ERPJEWELS

Latest Trends

ETL/Trends in Data mining/ Trends in Data mining/ Data tranformation and multi-dimensional repoting capabilites

Download

Docs, Executable, AVI and others in Download Section

Hot Links

Find what is Hot in ERPJewels

Related Links

PUSH INTELLIGENCE

Traditional PULL Intelligence

Traditionally intelligence information in any ERP is manual and user driven. User has to EXTRACT (collect) data, to TRANSFORM (calculate, sort, summarize) the same and to LOAD (organize) the same in the format desired. Further this has to be done every time he/she needs the same information for different period.  Thus intelligence was pulled by user and was heavily dependent on users and system could not push the same to users automatically. Thus analyses were user triggered rather than time triggered.

Job Management as instrument of PUSH Intelligence

Job management is an instrument in Baan to automate certain tasks and make it time triggered instead of user triggered .This job management is usually run for batch mode updates , end of day update processions , rebuild session requiring longtime . This is time tested instrument in Baan to automate periodical procsses based on time without user intervention.

Problems using Job Management in Periodic Analyses

Normally majority of the analyses in any Busine are routine in nature and periodic. Baan job management can be easily used for this purpose. However there are 2 i ues to be addressed.

1. Unequal period lengths

Normal analyses periods like months/quarters etc have unequal number of days in each period. Due to these when the program are run in job mode (let’s say in night without any user), the ‘from date’ goes wrong. And hence the analysis results will go wrong. An example will illustrate this vividly.

There is a report for the month of January 2007. A job is created in 2nd February for this program with from date as 1st January 2007 and to date as 31st January 2007. This will run ok only for 7 months in year with 31 days. Other months with number of days unequal to 31 days will not work correctly. E.g. in March for analysis of February the ‘from date’ will be 29th January 2007 (33 days earlier) and to date will be 28th February 2007 (2 days earlier). It’s evident this analysis results will be wrong.

2. Only Date fields considered in Job management and numeric time fields not considered.

This again is an issue. Because many a time triggered analyses and programs in Baan have not date fields as input. Instead it has numeric fields like fiscal period, week number etc. The automatic scheduling of analysis should be able to consider this numeric time fields. Further there should also be a provision to address the year change automatically.

Due to these factors analyses still remains user triggered and not time triggered.

Does all periodic analyses go wrong with job management

No. Not at all. In fact Baan job management is time tested process scheduling mechanism. It will be correct in following cases.

1.When the analysis period (based on date fields) has same number of days all the time e.g. weekly analysis.

2.When the report is based ‘as on certain date’ and not ‘from-to’ date. In such a case ‘as on date’ has to be period end related and job has to be created immediately in the next period. Or the job when the job is created in the same period in which ‘as on date’ exists.

How this is addressed.

As detailed above, the problem is functionally wrong ‘from date’ and non-recognition of non-date time fields by Baan job management. This results in wrong selection criteria when run in job mode. This in turn means wrong analysis results. So the solution will be to correct the from date before the Baan job is fired and to recognize the non-date time fields and it’s period type and modify these non-date numeric  time fields just before the job is fired.

1. Analysis Based on Baan Report

As Baan report is the output of Baan session. Baan session (selection criteria) determine the way the job will run. Any mechanism to schedule analyses based on Baan report seamlessly must consider job management mechanism and synchronize the same with Baan job management.

There are three different ways for creating job for running Baan report.

1.Printing reports using rprt_open() function in Baan in cont.process choice. Window similar to following on may or may not come while creating the job in session generating report (Tools è Make Job)

2.Report is printed using brp.open function. The reports and analyses created using spool functions will not come within the preview of this tool for automation.

3. Where report is selected through user.choice  ( not cont.proce).

Following code has to be inserted in the script if selected through button.

#pragma used dll otudllanl000001 for declaring dll used by this report

   Choice.cont.process

   Choice.cont.process:

   On.choice:

        if job.process then

             execute(execute.user.choices())

        endif

The following steps are to be performed to make an analysis based on report time triggered.

a. Create Job with appropriate field selection values. Also decide the analysis frequency.  If certain actions are to be done before running the job , ‘job waiting after execution’ may not be selected.

b.Following question will occur at time of ‘Make Job’. The user can answer ‘yes’ provided he/she runs the session tudll9166m000 on the same day for creating period offsets.

c. While creating the job give the appropriate sequence number. Please leave 1 sequence number extra to accommodate another job in between. e.g. If the last session no. in job A is 0 , then give this job number 2 instead of 1. This will enable to insert 1 job later on just before this job (to correct sequence selection parameters). In cases where the report is run through special choices (user.0, user.1 etc) as mentioned in case 3 above, the field information has to filled with user choice no. e.g. the report is coming through choice user.0 , this field has to be filled like ‘0 Stone Shape’ . Thus information in such case should start with the user choice number. Further when the job is created for consolidation later on , the field information (ttaad501.info) should contain word ‘consol’ anywhere in the field.

Record session, report and job details in following sess ion name and user does not have to key in. In cases where job management does not recognize the report while creating job (case 2 & 3 mentioned above) , the field report has to be kept blank (tudll9166m000).

Field

Description

Session

Baan session in which job is created.  User can zoom to get the available jobs.

Report

The report which is going to run in job mode. This field has to be left blank for cases 2 and 3.  User can zoom to get the available jobs.

Company

Company for which job is created. By default the current company.

Job

Baan job in which the session added. Default will come depending on the selected parameters. This will also be validated against valid job in Baan

Job sequence

Default will come for the selected values at the time of insert. It will be validated against job sequence in Baan

Job session no

Default will come for the selected values at the time of insert. It will be validated against job session in Baan


Button

Description

Job offset

It will invoke a new session (tudll9166s000) wherein user can define the job offset type.

Link analysis

When there are many analyses on a reports attached to the session , and user wants to run only specific analysis in job then user can selectively link analyses to the job.

 

          This session automatically populate the date fields for determining the job offset. Further it will link all analysis available for that report in current package combination automatically. Subsequently user can change this link.

 

e.      Select ‘Job Offset’ to provide job offset details. In the session tudll9166s000, all date fields will come by default.

Maintain the job offset depending on analysis frequency. Further non-date numeric time fields can also be maintained here to indicate analysis behavior along with year offset variable, if any. Period nature (as on date for period/from-to period) indicates whether to period variable is required or not. To period variable will be required when the period nature is from to period. Further in case of non-date time period variable and period nature – as on date/for period, year offset also can be specified.

g.      As one report can have many analyses linked to it, user can specify which all analyses to run when this particular job is run (tudll9560s000). By default, all analyses for the report / session (when report is blank) will be considered. Except for the cases where the report is printed through brp.open, all analyses available for the report / session concerned will be linked with job by default. The ones created with brp.open needs to linked by marking the analysis record. The field ‘Selected’ will be ‘Yes’  and highlighted if it’s linked to job or else it will be ‘No’.

Create job in tudll9166m000 (in the same job as the original job with session number lesser than the number for original report and session). Skip on error may be kept 1 to skip the next session.

Now the job is ready to run periodically in the background with correct selection parameters under job daemon.  The day on which job is to be run job daemon has to be activated (ttaad5206m000).

2. Analysis Based on Baan Table and ASCII file.

This is relatively simpler because the selection criteria are outside Baan standard. So when the job is created, job offset data and period frequency is determined. And at the time they are run in job mode, these details are used for determining the selection criteria. When running these data source for execution in job mode , care should be taken to write word ‘consol anywhere in field information (ttaad501.info ) so that consolidation program running afterwards will consider this analysis for consolidation.

Sophisticated Push Intelligence

It will be very evident that with the use of Baan Job management, coupled with period intelligent programs and ETL methodology of the tool, PULL intelligence will transform to PUSH intelligence thus saving lot of time, money and effort for the business which in turn lead to booming bottom line. This adds much needed vitality and agility to the business.



FastTrack to ERPJEWELS :

Latest Trend | Hot Links | Jewels of ERPJewels | Jewels of Baan | Jewels of Excel