Using bills of material to solve business problems
Another article on this website discusses the different types of bills of material (BOMs) that are available in full-function ERP systems, namely:
- Production BOMs
- Phantom BOMs
- Planning BOMs
This article looks at how properly structured BOMs can:
- Reduce stocks (of both raw materials and work in progress)
- Reduce production lead times
- Reduce delivery lead times to customers
- Reduce the BOM maintenance workload
- Improve BOM accuracy
Different ERP systems offer differing functionality, with some offering things like lead time offsets at a component level in the BOM so that the Materials Requirements Planning (MRP) module can see that not all items on the build are needed on the same day.
So when production lead times are measured in days, the system can be told that, for example, the packing items are not required until the last day. But we'll assume that only standard functionality is available. That means that MRP will schedule all necessary items to be available at the beginning of the production process: i.e. the works order release date.
For items that have short processing times, that is not a problem but if a particular works order is open for several days, issuing all of the items required before they are needed, often leads to log jams on the factory floor. That causes companies to leave items in the stores or warehouse until they can be used and, although this reduces work in progress (WIP), it has the direct opposite effect on raw material and, potentially, sub-assembly stocks.
And if the items are bulky, or have limited shelf lives, the problem is exacerbated. The answer can be to restructure the BOM so that each phase of production has its own works order, but a balance between too few and too many levels has to be achieved, as the relationship between raw materials and WIP is close and a reduction in one can cause an increase in the other.
However, consider an item with four production phases, each of one day. With a single-level BOM, all the materials required will be planned to be there on day one even though some (e.g. packing items) may not be needed until day four. Having four levels on the BOM means that the materials will be scheduled to arrive 'just in time', thus minimizing both raw material stock and WIP.
Reducing production lead times
Reducing production lead times is not only a good thing in itself, but has a major impact on WIP levels because Little's Law, amongst other things, tells us that WIP = Production rate x Lead time. Production lead time is defined as the elapsed time between starting a job and finishing it, so it is not the same thing as delivery lead time, which is defined as the elapsed time between receiving a customer's order and delivering it.
The difference lies in the lengths of the company's order book: if a company can produce 1000 items a week and has an order book of 10,000 items, then reducing the production lead time from one week to one day will have a minimal impact on delivery lead time.
Although having too few levels on BOMs can impact stock and WIP levels, having too many can affect WIP levels and production lead times. In a previous example, a product had four manufacturing stages, each of one day, and the MRP module of an ERP system will plan those operations to happen over four consecutive days, giving a 4-day production lead time.
But, if those operations take 1.1/2 days each (for a total production time of 6 days), that can be a problem because almost all MRP systems work in daily 'time buckets' and in consequence, the first operation will be planned for days 1 and 2, the second for days 3 and 4, etc.
That means that the job will be planned over 8 days, and not 6, increasing WIP time by about a third (and Little's Law tells us that WIP volume (and value) will increase by about a third also). The answer is that all full-function ERP systems come with a finite capacity scheduling option, which will bring the manufacturing lead time back down to 6 days (and in some circumstances even less).
However, some companies shy away from such systems because they need accurate information and properly trained people to perform optimally.
Reducing delivery lead times to customers
As has already been mentioned, reducing production lead times often has a minimal effect on delivery lead times (although it is still worth doing to reduce WIP levels), but ERP can still help. One way that manufacturers that have sub-assemblies in their BOMs can significantly reduce delivery lead times is to have products already part-made when customer orders arrive.
One manufacturer of PCs and servers used the following method to offer the next-day dispatch of any order received by midday. Key to this is the ability to forecast demand for individual sub-assemblies with a reasonable (though by no means 100%) level of accuracy.
The forecasts are then processed through planning BOMs to give forecast demand at sub-assembly and raw material levels. With a small amount of buffer stock, the company has a high level of confidence in having sufficient materials on hand when daily customer orders arrive.
If demand for a particular item is less than expected, MRP automatically nets off excess stock against replenishment demand on subsequent days so that the company never finds itself over-stocked. This all works for them because, although they have a wide range of products, they find that demand at the sub-assembly and raw material levels is surprisingly steady for most items and using this method frees up time for them to focus on exceptions.
Reducing BOM maintenance workload and increasing accuracy
The other article discusses the use of phantom BOMs to 'logically group' items that appear on many bills, with packing items being used as an example. Some companies producing electrical or electronic items can find that many items, from power cords to shrinkwrap, can be grouped in this way, as can shopfloor consumables such as staples, fasteners, and resistors.
That enables 'groups' of items to be called upon to BOMs rather than having to remember each item individually (and potentially forgetting one). But intelligent use of regular BOM functionality can make a significant difference to both maintenance workload and accuracy before items go into production, during the design phase.
When a new product is being developed, it can take a considerable period to complete the design and to create a production BOM that can be used by MRP to generate purchase orders for the necessary materials. Most textbooks and most consultants say that BOMs should not be entered into the system until they are complete.
But most textbooks and most consultants can sometimes be wrong. The problem is that MRP cannot generate purchase recommendations without a recognized demand. Some allow purchase requisitions to be raised but many companies try to manage demand via manual systems such as spreadsheets, and inevitably things fall through the cracks.
When manual systems fail (and they frequently do), the purchase of necessary materials is delayed until the completed BOM is read by MRP. That may not be a problem but, if some items are on extended lead times from suppliers, it can certainly be, because then the first time that manufacture can take place cannot be within the purchasing lead time.
So what can ERP users do to manage the situation? Well, they can use the functionality that their ERP system offers. The module of an ERP system generates recommendations to buy and to make items, but there is no reason that it can't be used to generate recommendations to make decisions.
The first step is to identify the types of decisions that are likely to be needed, such as customer decisions, marketing decisions, production engineering decisions, etc, and then to consider if there is value in breaking these down further. Customer decisions, for example, could be broken down into decisions on packaging, color, material, etc (each company will have its own needs and opinions).
Once established, each 'decision type' can be added to the product file (it is a good idea to give them a prefix, such as a 'Z' to keep them away from regular parts), and the lengths of time before the product launch or product delivery that decision needs to be taken by should be entered as a lead time.
For example, if it's likely that the packaging will be on a 30-day lead time once ordered, the decision on packaging needs to be taken at least 30 days before it is required. Now, when an incomplete BOM is entered into the system, outstanding decisions (codes) are entered into it.
Two things then happen. One is that every time MRP runs, it will recommend actions not only for purchase orders and works orders, but also on decisions that are coming due. And, secondly, all BOM modules have a 'where used' report that shows which BOMs still have decisions outstanding.
So items can't be forgotten and BOM accuracy improves. That's a lot of advantages from an under-appreciated module!
Featured white papers
ERP Software Pricing Guide
Get the latest pricing information on over 80 popular ERP systems, and learn how to budget for your ERP project in our free guideDownload
60-Step ERP Selection Checklist
Get the comprehensive checklist for your ERP selection projectDownload
ERP Implementation: 9 steps to success
The 9 proven steps you should follow when implementing ERPDownload
Using ERP for a competitive advantage
How ERP software can help you get ahead of the competition
Why a food specific ERP system is a must-have
Key features and requirements food companies should consider when searching for an ERP
Your ERP system may have more BOM functionality than you think
Our look at the different types of bills of material, and how they can help.