A few weeks ago I blogged about the unique warehousing needs of rolled and continuous products like cold and hot rolled metal coils and sheets, bales, paper rolls, cables and wires.  One of the drawbacks of using WMS in this type of an environment is that WMS does not recognize over picking of items and does not over-allocate more than ordered quantity to optimize picks.

Let me illustrate this shortcoming with an example e.g. if you have an order for a wire for 100 meters with a 10% over ship tolerance.  Assume you have bobbins in the warehouse with 105 meters, 110 meters and 150 meters of wires respectively. If you model the bobbins as LPNs and have full LPN allocation mode turned on, WMS will backorder the entire quantity even though two of the bobbins i.e. bobbins with 105 meters of wire and 110 meters of wire is within the tolerance.  If you do not have full LPN allocation mode, it will assume that you will cut 100 meters from one of the bobbins and allocate the rest to some other order.  This is even worse since it will unnecessarily force you to cut material when nothing like that is needed.

One of the customizations we did a long time ago for one our customers was to make sure that WMS looks at LPNs greater than the ordered quantity for allocation purposes. This works very well for the situations we illustrated earlier. In the above example, with this customization it’s possible to allocate more than the ordered quantity and create a task to pick the bobbin with 105 meters of wire.

The good news is that this feature is now available in the standard product. The feature dubbed “over allocation” in R12.1 works exactly like I outlined above. With over allocation mode, the task quantity can exceed the quantity ordered provided the overall allocated quantity is within the shipping tolerance. You can either specify the over ship tolerance within the rule or instruct the rules engine to use the over ship tolerance at order line level.  To take full advantage of this feature, you need to have full LPN allocation. The advantage of this feature is that the LPNs that contain more material than the ordered quantity, but within the over ship tolerance, are considered for allocation. There is no more issue of backorder and the need to cut products to get the desired size is minimized.

Please note however, this works only for Sales Orders at current time, not for orders from manufacturing where this is a common problem as well. We faced this issue while architecting the WMS solution at Zebra and have employed a different strategy to overcome this.

Rolled and continuous products have some unique characteristics that are particularly troublesome from a warehousing perspective. In this post I am going to talk about some industry needs the way I see them from my perch as a WMS implementer.

What are Rolled Product and Continuous Products?

You encounter these products every day. Think wires, cables, hot or cold rolled metal sheets, fabrics and paper rolls. Some of the unique challenges that one faces while dealing with these products are:

1. The items are planned, valued and priced based on weight or length. In other words, weight or volume unit of measure is most naturally suited for planning, manufacturing, valuing on-hand inventory and invoicing needs. However the items are physically stored as discreet units in the warehouse.
2. No standard packs. Due to the uniqueness about their production methods you are more than likely to get variable lengths of the product in your inventory. The variable nature poses problems as you cannot have a standard order size and as a result, the warehouses will result in smaller cut quantities of wires, rolls and sheets that need to be consumed somehow.
3. It’s difficult to fulfill an order exactly as ordered. Some of these items such as metal sheets or wires require effort to cut and therefore fulfilling an order exactly as ordered is difficult. Over or under picking of orders is the norm rather than exception for these products.

Over the years Gaea Global has implemented Oracle WMS at companies such as primary metal and utilities where we have comprehensively addressed these needs. The good news: most of the product functionality is available in the product and very few customizations are needed to address the major warehousing needs.

Let’s look at some of the WMS best practices from an implementation perspective:

1. Weight or Volume as the primary UOM: Since the items are planned, valued and priced based on weight or length, the primary UOM needs to be weight or volume.  The 11.5.10 catch weight functionality is not your friend as catch weight implies that you keep a discreet UOM as primary and capture quantity in a secondary UOM for invoicing needs only.
2. Define each discreet unit as LPN: This implies that every roll or bobbin is an LPN. This way you can store the quantity in LPNs without the need to measure or enter quantity every time a pick is performed.
3. Maintain accurate weight/volume for LPNs: This means from production to storage the LPN (i.e. roll, bobbin, bale, etc) must have accurate weight. This means that you do not need to weigh or measure the volume during pick time. If you happen to cut the product, the weight or volume of the product in the LPN must be updated.
4. Full LPN Allocation: Full LPN allocation ensures that residual quantities in LPNs get allocated and consumed. The out of the box allocation may not be most optimal and customizations may be needed to strike the right balance between quantity picked and quantity picked above tolerance.
5. Capture over ship tolerance: This needs to be captured at the time of order entry or defaulted for each item. Over ship tolerance ensures that item ship quantity is controlled.
6. Configure Pick with no over pick or under pick tolerance: Certainly you do not want to enter exceptions every time picks over or under. As I mentioned earlier, over or under picking of orders is the norm rather than exception for these products. You do want your operators to pick and confirm products without keying exception code every time.

What WMS does not do well is to recognize that over picking is common in the warehouse for such products. Therefore the tasks created will be consistently lower than the actual ordered quantity.

Daniel Soosai of Gaea will be co-presenting with Abhijit Kakhandiki, Director, Oracle Development at the Collaborate’10 on “Maintain Your Edge: Portfolio Management with Agile & Primavera” at the Mandalay Bay Convention Center, Las Vegas on April 20, 2010 from 8:00 a.m – 9:00 a.m
This presentation highlights cutting-edge product portfolio management strategies that can be enabled using Oracle’s Agile and Primavera and Gaea’s Agile-Primavera PIP.

Daniel Soosai of Gaea will be co-presenting with Murari Ram, Operations Manager, Cisco Systems at the Collaborate’10 on “In-search of Perfect Order Performance: Synchronizing Supply Rapidly with Demand & Promising Orders” at the Mandalay Bay Convention Center, Las Vegas on April 20, 2010 from 8:00 a.m – 9:00 a.m
This presentation provides an innovative solution and business process that could be used to improve the utopian supply chain metric – perfect order.

I recently watched a Webcast by Dan Gilmore about Labor Management. Dan essentially belabored the point (no pun intended): labor management is important for your warehouse. I will echo one of the points that Dan made in the webcast: labor management is much more effective in driving efficiency if you do predominantly piece picks.  It’s less effective in full case or pallet zones.  The reason is straight forward. Much of the labor savings in a warehouse is realized when you reduce the travel time and piece pick has a significantly more potential to cut travel time than case or pallet picks.

Not many are aware that Oracle WMS actually has a labor management feature. Obviously the usability of this product can be somewhat better but it does what it’s supposed to do Here are some tips to use WMS Labor Management system effectively:

1. Define Labor Standards: This is the basis of any labor management system and Oracle ‘s system is no different. If the labor standards in the warehouse do not reflect reality, the resulting report will be garbage too. If you don’t want to spend time and effort on getting labor standards using time and motion study, simply use historical averages. WMS lets you do that.
2. Keep it simple: Oracle labor management system allows you to define labor standards against multiple parameters e.g. you can define standard pick time and travel time to perform case picks in a refrigerated zone. My suggestion: keep it simple. Don’t attempt to define complicated labor standards e.g. standards for conveyable products in a floor zone for different pick equipment types. There are two reasons I am cautioning against overly complicated standards. One, the user interface is not designed for complicated setups. Second, you risk entering incorrect standard times for unusual pick configurations.
3. Focus on piece pick area: As mentioned before, you will get more efficiency gains from your each or piece picks. Therefore focus on entering more detailed and accurate information on piece pick zones.
4. Focus on picks: Forget putaways or replenishments. Most productivity gains will be accrued in your outbound picks. Focus on picks.
5. Don’t try to use it as a tactical tool: Labor Management in Oracle WMS works well if you are interested to know average picker productivity and especially the trend over time. It can also be used to benchmark your pickers against a standard. However it’s possible, it is not a good tool for daily or wave level load balancing or using it for day-to-day resource planning. Wave planning gives you greater flexibility to plan wave level labor and resource requirements. Unfortunately the two systems use different sets of data even though they relate to essentially the same thing.

Last thing I want to point out is that, you do not need to be restricted to just the out of the box reports available in labor management. The greatest advantage is that labor management stores the most granular labor and task execution information. The rich data can be very easily used to generate customized reports and analytics.