KPI: scrap
Scrap: The output of a production process that is not finished goods or WIP (work in progress). Scrap usually costs money and time.
Scrap is usually a waste of:
- Materials– usually you cannot recycle the material, but even if you can, most of the time the recycling process produces scrap too, or some of the materials are contaminated and cannot be recycled.
- Rework time– the scrap that is being produced takes the machine time that produces the goods. For example, if a machine produces 80% goods and 20% scrap, by reducing the scrap to 12% we can get 10% more goods output at the same operating cost.
- Energy– when we produce scrap, we waste energy. When we recycle, we use twice the energy to get the same result.
- Handling– it takes a lot of effort to handle scrap, It needs to be moved from the machine to the storage location. From there to the recycling facility. Then back again to the storage location as raw materials.
- Disposal cost – sometimes when scrap cannot be recycled we need to pay someone to remove the materials from the factory.
- Quality assurance cost– if the production process creates a lot of scrap, then the quality assurance process should be tighter and that means more manpower and resources. For example, If your process produces 1% scrap, you need much less quality assurance resource than if you have 30% scrap if you want the customer to have the same positive experience.
Usually we can find 3 major scraps (we need to separate them because we fix them differently):
- Bad products– material/WIP that was transformed to a product, but is not in the QA spec.
- Rework– bad products that need minor adjustment to become good products.
- First Material– every time the tool starts it usually consumes materials and creates off spec materials due to unadjusted parameters, The tool needs to get warmer, The first material “cleans” the machine, and so on.
Cost of scrap could be calculated in one of the following ways:
- Calculating the actual cost needed using the 6 wastes above. This is not easy to calculate and needs many assumptions. This type of calculation is very important to understand why we need to allocate resources for QA.
- Calculating the cost of the finished goods.Since each piece of finished goods has a value as inventory, we can give scrap the value of the finished goods. Example: if we produce finished product XXX that has a value of $100 per unit/ton, then if we produce 1,000 products and 200 pieces of scrap, we produced $100,000 of goods and $20,000 of scrap. This method is very easy to calculate and allows us to use % of scrap from the total batch.
Since there are 3 scrap categories, we have many possible graph variations:
1. Scrap (quantity) stack bar by month
Short description: the number of scrap produced each month (even if it was recycled later)
Who uses it (importance): higher management (High), factory management (High), unit management (High)
Description: A stack bar of all the actual scrap quantity produced each month divided to categories (Bad products, Rework, First material). That shows the actual quantity of scrap that the factory needs to deal with.
2. Cost of scrap ($$) stack bar by month
Short description: the cost of scrap produced each month (even if it was recycled later)
Who uses it (importance): higher management (High), factory management (High), unit management (High)
Description: A stack bar of all the cost of scrap produced each month divided into categories (Bad products, Rework, First material). The KPI shows the actual cost of scrap that the factory needs to deal with. It is a very important figure since it determines the resources given to improve QA and production. Another variation of this graph can be a show of total scrap cost (without the categories).
3. Percentage of scrap
Short description: the % of scrap from all input materials
Who uses it (importance): higher management (High), factory management (High), unit management (High)
Description: the % of scrap from the entire material input is very important to understanding how much money we lose in the production process. It can be calculated to each step, to the overall steps, and broken into the basic scrap categories (Bad products, rework, first material).
The basic formula:
- % of scrap = Scrap material / Materials intake (amount of material put into the process)
4. Scrap cost vs. % of scrap – current month
Short description: the cost of scrap for each tool this month vs. the percentage from the total intake
Who uses it (importance): higher management (Low), factory management (High), unit management (Med)
Description: The cost of scrap in $ vs. the percentage of scrap in kg is a very important KPI. It allows us to understand which tool contributes more to the scrap cost. The % of kg tells us how serious is the problem. When initializing a cost reduction process we should start investing time and money according to the table:
We need to start reducing the high cost of scrap and the high percentage because there we have more potential and it will contribute more to cost reduction. After that we need to look for low percentage and high cost. Usually it means that this tool uses more expensive materials or very high volume. In that case every small change that we will make will reduce the cost of scrap. In the following graph we can see that tool 27 and tool 15 already did a scrap reduction process (both have a low %, under 3%), but still have a very high cost of scrap. That is why they need to be constantly under a cost reduction process. We know that every reduction of 10% of the scrap they produce will contribute very much to reducing the scrap cost. However, tool 16, 17, 5, and 6 can easily be reduced to 50% of their scrap which will also contribute to cost reduction. The 4 tools will have much better success rates by implementing what we did with tools 27 and 15, which is why we need to start with them.
Note: this graph can also be very effective if we use accumulated 6 months’ data.
Design tip – sort the tools from high cost to low cost. That will give a clear view of the problem.
5. Scrap per tool – current month
Short description: scrap per tool current month (one graph for each scrap category)
Who uses it (importance): higher management (Low), factory management (Med), unit management (High)
Description: The following graph is used for the tool owners and for any cost reduction process manager. It gives the actual data of each scrap. In most cases the owner will look for reducing one type of scrap at a time.
6. Scrap per process stage
Short description: the percentage of scrap generated by each process stage
Who uses it (importance): higher management (High), factory management (High), unit management (High)
Description: If the production process has multiple stages, we need to understand how scrap is distributed across the process stages. We can look at the stages in terms of kg/units if it is relevant to all stages, or in terms of cost. The cost perspective allows us to give different prices to scrap at different stages. For example, if we extrude something, then paint and wrap it, we need to give different costs to each step. We lose X$ per unit at the extrusion stage, 130%X $ at the paint stage (we already invested time and material to extrude and to paint), and 150%X $ at the wrapping stage. Then we calculate all the costs and convert them to percentages. The scrap per stage cost table will look like the following example:
In the above table we can see the stage and in detail how much goes to every scrap type
And then we can go even deeper:
6. Monthly average % of scrap quantity vs. cost
Short description: comparing the monthly average percentage of scrap in terms of cost vs. quantity.
Who uses it (importance): higher management (med), factory management (med), unit management (med)
Description: If we take the scrap cost only as the cost of materials, we will get the same percentage per batch both in quantity and in quality. But when we look at an average, if one number is higher than the other, it explains the type of scrap. If the cost percentage is higher than the quantity (as in the example), we lose more expensive materials which means we lose more materials on expensive product batches than on cheaper ones. It might be because they are more sophisticated to produce. In this case we should first look to improve tools that run more expensive materials.
