B Appendix: RDF Cloud Service integration with RMF Cloud Service

This appendix describes RDF Cloud Service integration with RMF Cloud Service

Integration Assumptions

Following is a list of requirements/assumptions for integration with RMF CS:

RMF CS sends hierarchy and data files on a weekly basis. Hierarchy files are sent as full set. The data files are sent incrementally.
RMF CS sends the data at item level.

Integrated Hierarchy and Data Files

RMF CS sends the following sets of hierarchy and data files. All RMF CS files are copied to the common cloud service share location from where RDF CS can pick up the files for further processing. Interface files need to be transformed to format the files into the RDF CS required format to filter columns not used and split and create different group of files as needed by RDF CS. RDF CS Configured Batch Process will transform the files to the RDF CS required formats, rename the final files as needed by RDF CS, and copy them to the domain input location for subsequent hierarchy or measure load processing. For each of the hierarchy and data files, RMF CS also sends a trigger file with the same name as the original file with the extension .complete~cloud. The RDF Cloud Service batch process waits for the trigger files to start processing the corresponding data or hierarchy file. After processing, it deletes those trigger files.

Hierarchy Files

The following hierarchy files are provided by RMF CS and need to be transformed to the RDF CS format before loading them into RDF Cloud Service. For each hierarchy file, RMF CS only sends the base and alternate dimensions of the hierarchies and not the user alternate rollups.

Calendar Hierarchy - rms_clnd.csv.dat

RMF CS sends the calendar hierarchy file without calendar labels. Calendar labels are added in the weekly batch process by calling the RDF CS custom script (ra_custom) with the parameter rms_rdf_clnd_label, which adds the calendar labels.

RMF CS sends the calendar hierarchy file with dates for five years. Based on the current date, two year historical dates, current year dates, and two future year dates.

Table B-1 contains the list of column contents from the file. Only required columns for RDF CS are filtered by the transform process.

Table B-1 Calendar Hierarchy File Fields

Field Name	Field Description	RDF CS Mapping
Day	The date from which the 4-5-4 data was derived, in YYYYMMDD format.	Day
Week	The end of week date for the day, in YYYYMMDD format.	Week
Month	The 4-5-4 month of the year, valid values 1 to 12.	Mnth
Quarter	The 4-5-4 quarter of the year, valid values 1 to 4.	Qtr
Half	The 4-5-4 half of the year, valid values 1 or 2.	Half
Year	The 4-5-4 year.	Year
week_of_year	The 4-5-4 week of the year, valid values 1 to 53.	WOYR
day_of_week	The day number within the week, valid values 1 to 7.	DOW

Note:

When the RMS calendar extract is imported into RDF, the RMS month id 1 corresponds to February in RDF. It follows the 4-5-4 Retail Calendar. RDF has provided a numeric domain property app_start_month to override the default behavior. This can be done using OAT.

The default value of app_start_month =2, means RMS month id 1 will correspond to February.

If app_start_month = 10, means RMS month id 1 will correspond to October.

If app_start_month = 10, means RMS month id 12 will correspond to September.

Product Hierarchy - rms_prod.csv.dat

RMF CS exports all sellable and inventoried items.

Table B-2 contains the list of column contents from the file. Only required columns for RDF CS are filtered by the transform process. Some measure data files are also extracted and loaded from this RMF CS product hierarchy file.

Table B-2 Product Hierarchy File Fields

Field Name	Functional Name	RDF CS Mapping
ITEM	Item ID	sku
ITEM_DESC	Item Description	sku_label
ITEM_PARENT_DIFF	Parent/Diff ID	skup
ITEM_PARENT_DIFF_DESC	Parent/Diff Description	skup_label
ITEM_PARENT	Parent ID	skug
ITEM_PARENT_DESC	Parent Description	skug_label
SUBCLASS_ID	Subclass ID	scls
SUB_NAME	Subclass Name	scls_label
CLASS_ID	Class ID	clss
CLASS_NAME	Class Name	clss_label
DEPT	Department	dept
DEPT_NAME	Department Name	dept_label
GROUP_NO	Group	pgrp
GROUP_NAME	Group Name	pgrp_label
DIVISION	Division	dvsn
DIV_NAME	Division Name	dvsn_label
COMPANY	Company	cmpp
CO_NAME	Company Name	cmpp_label
FORECAST_IND	Forecastable Item flag	Not Applicable
CLASS_DISPLAY_ID	Class Display ID	Not Applicable
SUBCLASS_DISPLAY_ID	Subclass Display ID	Not Applicable
BRAND_NAME	Brand ID	Not Applicable
BRAND_DESCRIPTION	Brand Description	Not Applicable
SUPPLIER	Supplier Site	vndr
SUP_NAME	Supplier Name	vndr_label
DIFF_TYPE1	Diff Type1	Not Applicable
DIFF_ID1	Diff 1	Not Applicable
DIFF_TYPE2	Diff Type2	Not Applicable
DIFF_ID2	Diff 2	Not Applicable
DIFF_TYPE3	Diff Type3	Not Applicable
DIFF_ID3	Diff 3	Not Applicable
DIFF_TYPE4	Diff Type4	Not Applicable
DIFF_ID4	Diff 4	Not Applicable

Location Hierarchy - rms_loc.csv.dat

Table B-3 contains the list of column contents from the file. Only required columns for A&IP FSL CS are filtered by the transform process.

Table B-3 Location Hierarchy File Fields

Field Name	Functional Name	A&IP CS Mapping
LOCATION	Location ID	stor, strc
LOC_NAME	Location Name	stor_label,strc_label
DISTRICT	District ID	dstr
DISTRICT_NAME	District Name	dstr_label
REGION	Region ID	regn, tdar
REGION_NAME	Region Name	regn_label, tdar_label
AREA	Area ID	chnl, chnc
AREA_NAME	Area Name	chnl_label, chnc_label
CHAIN	Chain ID	chan
CHAIN_NAME	Chain Name	chan_label
COMPANY	Company ID	comp
CO_NAME	Company Name	comp_label
COMPANY_CURRENCY	Primary Currency	Not Applicable
LOC_TYPE	Location Type	Not Applicable
LOC_TYPE_NAME	Location Type Description	Not Applicable
PHYSICAL_WH	Physical WH ID	Not Applicable
PHYSICAL_WH_NAME	Physical WH Name	Not Applicable
CHANNEL_ID	WH Channel ID	Not Applicable
CHANNEL_NAME	WH Channel Name	Not Applicable
STORE_CLASS	Store Class	stcl
STORE_CLASS_DESCRIPTION	Store Class Description	stcl_label
STORE_FORMAT	Store Format	sfmt
STORE_FORMAT_NAME	Store Format Name s	fmt_label

Product Attribute Hierarchy - rms_uda.csv.dat, rms_brand.csv.dat, rms_vendor.csv.dat, rms_diff.csv.dat

UDA, Suppliers (Vendor), Brand, and Diffs in RMF CS are treated as product attributes in RDF Cloud Service. RMF CS provides them as individual files in the same format; those will be merged and loaded as one product attribute file. UDA_VALUE (Product Attribute Values) are not unique across UDA_ID in RMF CS. While transforming and loading as a product attribute, UDA_ID position will be concatenated with UDA_VALUE using an underscore ("_").

Table B-4 contains the list of column contents from the file. Only required columns for A&IP FSL CS are filtered by the transform process.

Table B-4 Product Attribute Hierarchy File Fields

Field Name	Functional Name	RDF CS Mapping
UDA_ID	UDA	Patt
UDA_DESC	UDA Description	patt_label
UDA_VALUE	UDA Value	patv
UDA_VALUE_DESC	UDA Value Description	patv_label

RDF Cloud Service Output Data

The following data files are provided by RMF CS and need to be transformed to the RDF CS format before loading them into RDF Cloud Service. All RMF CS files are transformed first and loaded together using the single Load Task Load - RMS Interface Data.

Sales Data File - wsales.csv.ovr

Table B-5 contains the list of column contents from the file. Only required columns for RDF CS are filtered by the transform process. RMF CS sends Net Sales in all regular, promotion, and clearance buckets. The transform process will split the sales into three files, rsal.csv.ovr, psal.csv.ovr and csal.csv.ovr based on the buckets.

Note:

The Other Sales (osal) measure has not been currently integrated with RMFCS and can only be loaded into RDFCS using a flat file.

Table B-5 Sales Data File Fields

Field Name	Functional Name	RDF CS Mapping
ITEM	Item ID	sku
LOCATION	Location ID	stor
EOW_DATE	End of Week Date	week
SALES_UNITS	Sales Unit	rsal,psal or csal
SALES_TYPE	Sales Type (regular, promotion or clearance)	Used to divided sales units into different files

Outage Indicator Data File - ldpreoosind.csv.ovr

Table B-6 contains the list of column contents from the file. Only required columns for RDF CS are filtered by the transform process.

Table B-6 Outage Indicator Data File Fields

Field Name	Functional Name	RDF CS Mapping
ITEM	Item ID	sku
LOCATION	Location ID	stor
EOW_DATE	End of Week Date	week
INDICATOR	Outage indicator	ldpreoosind

Item Attribute Mapping File - prdattt.csv.ovr

Table B-7 contains the list of column contents from the file. Only required columns for RDF CS are filtered by the transform process. This file only contains Product Attribute mappings for UDA attributes. Item Attribute mappings for Brand, Supplier, and Diffs are derived and loaded from the RMF CS Product file.

Table B-7 Item Attribute Mapping File Fields

Field Name	Functional Name	A&IP CS Mapping
ITEM	Item ID	sku
UDA_ID	UDA	patt
UDA_DESC	UDA Description	Not Applicable
UDA_VALUE	UDA Value	prodattt
UDA_VALUE_DESC	UDA Value Description	Not Applicable
FORECAST_IND	Forecastable Item	Only items that have the flag set to True are imported into RDF CS.

Integrations Between RDF Cloud Service and other Applications

This section details:

Batch Tasks Specific to RMF CS Integration
Implementation Steps for RMF CS Integration

Batch Tasks Specific to RMF CS Integration

The following sets of Online Administration Tools tasks are pre-configured in Configured Batch Tasks to support RMF CS integration. For more details about running the tasks, see the Oracle Retail Demand Forecasting Cloud Service Administration Guide.

Configured Batch Task > Run Batch Measure Export Group > Export Daily Demand

This task exports daily forecast and cumulative interval into a file named dsdemand.csv.ovr and push it to SFTP output and the common cloud service share location.
Configured Batch Task > Run Batch Measure Export Group > Export Weekly Demand

This task exports weekly forecast and cumulative interval into a file named wsdemand.csv.ovr and push it to SFTP output and the common cloud service share location.
RDF Cloud Service Batch Task > Run RDF Cloud Service Batch > Export Demand

This task performs the work of both Export Daily Demand and Export Weekly Demand.
RDF Cloud Service Batch Task > Run RDF Cloud Service Batch > Transform RMS File

This task only transforms all RMS hierarchy and data files to the RDF CS required format and copies the hierarchy files to the SFTP output as rms_hier.zip. It will not load those files.

Typically, this can be used during initial installation to transform the initial set of RMF CS files and rebuild the domain with the initial set of RMF CS hierarchy files.

Implementation Steps for RMF CS Integration

If RMF CS needs to be integrated as part of an initial implementation, then the following needs to be done in this order:

Install RDF Cloud Service using the Installer which will install the default solution with GA data files.
Upload the RMF CS Hierarchy and data files for the initial conversion data.
Run the Transform RMS Files batch task to transform hierarchy files to RDF CS format. Those will be copied to SFTP output.
Upload the generated hierarchy files and also additional hierarchy files needed for RDF CS, that are not provided by RMF CS (use the GA data set as a reference to create additional hierarchy files) into the SFTP server, at the INCOMING_FTP_PATH/input subdirectory.
Using the Self-Service Administration Task in the bootstrap domain, rebuild the domain with the required plug-in options for RDF CS and without selecting Use GA Data. The process will rebuild the domain using the initial set of RMF CS hierarchy files.
After building the domain, re-send all the hierarchy and data files from RMF CS and use the regular Weekly Batch to process the hierarchy and data files.