Fraska Portal

Exploring the WebSphere Commerce world

WebSphere Commerce, ATP migration

Posted by on in Store
  • Font size: Larger Smaller
  • Hits: 6088
  • 37 Comments
  • Subscribe to this entry
  • Print
WebSphere Commerce, ATP migration

When I started to analyze (summer 2015) the possibilities to migrate non-ATP stores to the ATP model the IBM Knowledge Center was claiming the migration was not supported. However, not supported does not mean not possible.

So, I dig it and found a "simple" way to accomplish the migration.

Now (since 2016), finally, the switch of an existing store to the ATP mode is supported again (old version of WCS had even a migration script).

In this brief article I'll report few tips about the "migration" of existing stores to the ATP inventory model.

Target:

WebSphere Commerce professional

Environment:

  • WebSphere Commerce Developer v 7.0.0.9 FEP8
  • WebSphere Application Server 7.0.0.31
  • DB2 9.7
  • SOLR 4.3.0 (local)

A simple flag, STORE.INVENTORYSYSTEM, determines the type of inventory the store manages. Of course, this configuration has an huge impact of the whole order flow and so the handling of the stock.

I started with an extended site store which had the non-ATP model; STORE.INVENTORYSYSTEM was -2. The aim was to switch this store to the ATP model.

The switch to the ATP consist on update this property to -1. However, there are multiple things to take in account since an ATP model is not so trivial to handle.

 

Fulfillment center

The value of the property FFMCENTER.INVENTORYOPFLAGS should be determined answering the following questions:

  • Will order accept multiple items? (multi-item)
  • Do you want inventory check operation always appear to succeed? (no-check)
  • Do you want the allocate operation to check inventory, but not allocate it (the allocate operation appears to succeed if the check succeeds)? (no-allocate)
  • Do you want the backorder operation to always fail so no inventory is ever backordered? (no-backorder)

 

Store

There are multiple properties should be filled in order to set properly the ATP behavior:

  • What’s the expiration time of the inventory allocation? STORE.ALLOCATIONGOODFOR
  • What’s the maximum number of days an item can be backordered? STORE.MAXBOOFFSET
  • What’s the default value of days an item can be backordered (to use when an estimated availability time cannot be determined for a backordered order item)? STORE.DEFAULTBOOFFSET
  • What’s the amount of time before the requested shipping date (ORDERITEMS.REQUESTEDSHIPDATE) when inventory allocation is performed? STORE.ALLOCATIONOFFSET
  • What’s the maximum time that can be requested for a future shipping date after the date when the requested ship date (ORDERITEMS.REQUESTEDSHIPDATE) is entered? STORE.MAXFOOFFSET

 

Scheduled Jobs

For the ATP WCS provides a set of Scheduled Jobs could be key to deliver properly the customer needs.

  • ReleaseExpiredAllocations:
    Releases order item allocations that exceed the expiration time. The scheduler reverses inventory with expired allocation for order items with status 'P','I','W','N', and inventory status is in 'ALLC' or 'BO'.
  • RAReallocate
    Reconciles expected inventory receipts to existing backorders to determine current expected ship dates. The command evaluates all backordered order items against current expected inventory receipts. If there are order items with an inventory status of 'BO', it reconciles to determine the expected ship date.
  • ProcessFutureOrders
    For locked orders in states 'C', 'M', 'I', 'B', 'L', 'A', and have some order items whose inventory status in 'FUT', this scheduler command allocates inventory for these order items, re-prepares orders, then process the orders
  • ProcessBackorders
    For the given store, get the list of locked orders that have a status of ('C', 'M', 'I', 'B', 'L', 'A') in the ORDERS table and an INVENTORYSTATUS of BO in the ORDERITEMS table. The command allocates inventory for these order items, re-prepare orders and then process the orders.
    For each of these orders:
    - See if there are any order items that are not validly allocated. (An order item is validly allocated if it is allocated, the quantity, fulfillment center, and shipping address are the same as when it was allocated, and the allocation has not yet expired.)
    - If there is at least one order item that is not validly allocated, call AllocateInventoryCmd for the order.
    - If an item is allocated from a different fulfillment center than the one originally requested and AllocateInventoryCmd has unlocked the order, then call ReprepareOrderCmd to recalculate shipping charges and taxes.
    - Call ProcessOrderCmd.
  • ReleaseToFulfillment
    Queries orders in the following states: 'C', 'M', 'I', 'B', 'L', 'A'. It checks whether the order is ready for release. If the order is ready to release to fulfillment, it changes the order status to 'R'.

So, according to the requirements above scheduled jobs should be configured and enabled or merely disabled.

 

ATP data model and data load

One of the highly impacting component is the data load since the ATP data model contains new tables and relationships to deal with.

In particular, the catalog data load should properly fill the following tables- for the correct work of the ATP process-:

  • BASEITEM
  • STOREITEM
  • ITEMVERSN
  • ITEMSPC
  • VERSIONSPC
  • STOREITMFFC

The stock have to be loaded now in the table RECEIPT and, according to the specific ATP model needed, there are new tables to fill

  • ATP inventory adjustment (additional tables involved INVADJUST, INVADJDESC, INVADJCODE),
  • ATP inventory expected (additional tables involved VENDOR, RA, RADETAIL, RABACKALLO),
  • ATP inventory fulfillment (additional tables involved ORDRELEASE, PICKBATCH, MANIFEST, ORDPICKHST, ORDSHIPHST),
  • ATP inventory orders (additional tables involved ORDRELEASE, PICKBATCH, MANIFEST, ORDPICKHST, ORDSHIPHST, BKORDALLOC, RABACKALLO, BKORDERITEM), and
  • ATP inventory reservations (additional tables involved INVRSRVTYP, INVRSRVDSC).

 So, the Inventory data load should be accurately planned and implemented.

Most of the effort will be probably spend of the configuration of the data load (to take in account for any eventual plan).

 

Front end and the expected date

Once a product have been configured as backorderable it will appear in the following way (in Aurora) if it's out of stock (no stock in RECEIPT):

b2ap3_thumbnail_2016-03-29_1322.png

Than, the checkout will show the customer when it will be available (the expected date):

b2ap3_thumbnail_2016-03-29_1323.png

In order to properly configure the expected date the data load should fill the tables RA and RADETAIL (as mentioned above).

If these tables are not filled the expected date is calculated according to the property STORE.DEFAULTBOOFFSET.

 

Some useful example of data load CSV files

Example of CSV non-ATP

CurrentStoreIdentifier,PartNumber,CatEntryStoreIdentifier,FulfillmentCenterId,FulfillmentCenterName,Quantity,QuantityUnit,Delete
Sample B2B Store,Cords-Black-29W x 28L,Extended Sites Catalog Asset Store,10101,Madisons Home,201,C62,0


Example of CSV ATP

PartNumber,ParentPartNumber,Type,INVQuantityMeasure,INVQuantityMultiple,TrackingInventory,BackOrderable,ReleaseSeparately,Creditable,ForceBackorder,ReturnNotDesired,MinQTYForSplit,PickingMethod,Discontinued,Delete
Cords-Black-29W x 28L,Cords,ItemBean,C62,1,Y,Y,N,Y,N,N,0,F,N,0

Example of CSV ATP (backorder enabled)

PartNumber,ParentPartNumber,Type,INVQuantityMeasure,INVQuantityMultiple,TrackingInventory,BackOrderable,ReleaseSeparately,Creditable,ForceBackorder,ReturnNotDesired,MinQTYForSplit,PickingMethod,Discontinued,Delete
GSH020_2008,,ProductBean,C62,,,,,,,,,,,
GSH020_200801,GSH020_2008,ItemBean,C62,1,Y,Y,N,Y,N,N,0,F,N,0

Note:

The ATP inventory configuration data are typically loaded together with catalog entry data in a single CSV file for initial loading, because each ATP inventory configuration data is related to a catalog entry record in the database. In this case, CatalogEntryMediator is used to load data.
In another case, catalog entry data are initially loaded by CatalogEntryMediator without ATP inventory configuration data, so default ATP inventory configuration data are loaded in to database. If user wants to modify the ATP inventory configuration data after that, FulfillmentPropertyMediator can be used to load only ATP inventory configuration data without impact to existing catalog entry data.

 

Example of CSV ATP (load the receipt)

PartNumber,CreateTime,ReceiptDate,Quantity,Cost,Currency,VendorName,Comment1,Comment2
GSH020_200801,2015-06-03 01:01:01.123456,2015-06-03 00:00:00.0,1,1,USD,eFEP8 Vendor,comment 1,comment 2

Example of CSV ATP (load the expected date)

VendorName,RACreateTime,PartNumber,ExpectedDate,Quantity,RADetailComment
eFEP8 Vendor,2015-06-02 11:36:01.123456,GSH020_200801,2015-06-28 00:00:00.0,10,radetail comment

 

Some useful query to check the ATP data

Query RECEIPT knowing the CATALOG_ID, STORE_ID and the PARTNUMBER

SELECT RECEIPT_ID, VERSIONSPC_ID, QTYRECEIVED, QTYINPROCESS, QTYONHAND, QTYINKITS FROM RECEIPT WHERE VERSIONSPC_ID IN
    (SELECT VERSIONSPC_ID FROM VERSIONSPC WHERE ITEMSPC_ID IN
        (SELECT ITEMSPC_ID FROM ITEMSPC WHERE BASEITEM_ID IN
            (SELECT BASEITEM_ID FROM STOREITEM WHERE BASEITEM_ID IN
                (SELECT BASEITEM_ID FROM CATENTRY WHERE CATENTRY_ID IN
                    (SELECT CATENTRY_ID FROM CATGPENREL WHERE CATENTRY_ID IN
                        (SELECT CATENTRY_ID FROM CATENTRY WHERE PARTNUMBER = '<PARTNUMBER>')
                          AND CATALOG_ID = <CATALOG_ID>)
                 )
             AND STOREENT_ID = <STORE_ID>)
         AND PARTNUMBER = '<PARTNUMBER>')
    )

 Query ITEMFFMCTR knowing the CATALOG_ID, STORE_ID and the PARTNUMBER

SELECT * FROM ITEMFFMCTR WHERE ITEMSPC_ID IN
    (SELECT ITEMSPC_ID FROM ITEMSPC WHERE BASEITEM_ID IN
            (SELECT BASEITEM_ID FROM STOREITEM WHERE BASEITEM_ID IN
                (SELECT BASEITEM_ID FROM CATENTRY WHERE CATENTRY_ID IN
                    (SELECT CATENTRY_ID FROM CATGPENREL WHERE CATENTRY_ID IN
                        (SELECT CATENTRY_ID FROM CATENTRY WHERE PARTNUMBER = '<PARTNUMBER>')
                          AND CATALOG_ID = <CATALOG_ID>)
                 )
             AND STOREENT_ID = <STORE_ID>)
         AND PARTNUMBER = '<PARTNUMBER>')

 

 

References

 

Rate this blog entry:
0
Tagged in: ATP Inventory Migration non-ATP WebSphere Commerce

Comments

Leave your comment

Guest
Guest Monday, 26 August 2019

Most Popular Post

WebSphere Commerce, the curious life of a front-end catalog request
Core
Rate this blog entry:
5
36 comment(s) 16966 hits Read More
on 08 October 2014
WebSphere Commerce, the SOLR extension index
Administration
Rate this blog entry:
4
52 comment(s) 15813 hits Read More
on 09 July 2014
WebSphere Commerce, Data Load and SOLR Delta Index
Data Load
Rate this blog entry:
0
128 comment(s) 14938 hits Read More
on 04 June 2013

Tag Cloud

Javacore Marketing Store Promotions Inventory non-ATP Search Expression Processor Content ClassNotFoundException Core Dynacache Search Navigation Flow RAD SOLR WebSphere Application Server CommandLevelAuthorizationCacheCmdImpl Index JR47273 DataLoad WebSphere Commerce European stores Validation Deploy Search Expression Provider Coupons Precision Marketing ATP Migration REST InitializationServlet Performances Toolkit Data cache preprocess WAS Plugin Congratulations Thank You Quality Management Center SoapUI Delta Deadlock Tables delta Landing page WCUserDistributedMapCache FEP7 Subcore WebSphere Commerce Developer Cores Data Model Web Activity Admin UI Troubleshooting Quality Assurance Team

Categories

Latest Blogs

WebSphere Commerce, CommandLevelAuthorizationCache
Cache
Rate this blog entry:
0
Comments (46) Hits (3369)
Posted by Francesco Schettini on 08 December 2016
WebSphere Commerce v8, toolkit exception, ClassNotFound db2
Administration
Rate this blog entry:
1
Comments (36) Hits (4607)
Posted by Francesco Schettini on 18 April 2016
WebSphere Commerce, ATP migration
Store
Rate this blog entry:
0
Comments (37) Hits (6088)
Posted by Francesco Schettini on 28 March 2016
WebSphere Commerce, the curious life of a front-end catalog request
Core
Rate this blog entry:
5
Comments (36) Hits (16966)
Posted by Francesco Schettini on 08 October 2014
WebSphere Commerce, Performance analysis of few European stores
Performance
Rate this blog entry:
0
Comments (27) Hits (4548)
Posted by Francesco Schettini on 22 July 2014