The standard deviation of demand during lead time is the same whether or not lead time is variable.

Try the new Google Books

Check out the new look and enjoy easier access to your favorite features

This is the second article in a series on setting safety-stock to address supply variation. Part 1 can be found here.

Analyzing Supplier Lead Time
To analyze supplier lead time, reference recent history to determine the typical number of days from order placement to receipt of goods. Remember, a day should be consistently defined the same whether conducting lead time analysis, demand analysis, or calculating kanban solutions. That is, if monthly demand is based on workday demand then lead time should also be defined in terms of workdays.

To get an accurate performance picture, we must analyze supplier lead time for each item and overall. Keep in mind that lead time must accurately define the current state as it actually is, not an ideal or pending future state. Far too many inventory managers fall into the trap of conflating lead time goals and actual lead time, and then knowingly or unknowingly trying to make up the gap with safety-stock. Don’t do this!

The purpose of supply-related lead-time safety stock is to mitigate the risks of actual supply variation. Thus, we must accurately quantify lead time variation. In that effort, 2 key metrics that give us a solid grasp of lead time variation are average lead time and lead time standard deviation. We must calculate both.

1. Average Lead Time: The average lead time is the mean of all actual lead times during a specific period of time.
   

   Use the AVERAGE function when calculating the standard deviation in Excel.

   Average Lead Time =AVG(LT1,LT2,LT3…LTi)

   In the above example, you would replace “LT1” with the number of days for the first lead time you observed. List lead times all the way up to the inth lead time you observed.

2. Lead Time Standard Deviation: Standard deviation is a measure of the distance that the typical data point within a specified data set tends to fall from the average of all data points in the same set. The farther the typical data point tends to be from the mean, the higher the standard deviation, which indicates more volatility. Thus, lead time standard deviation is a measure of lead time volatility for one or more related items.
   

   Use the STDEV.S function to calculate the standard deviation of a sample in Excel.

   Lead Time Standard Deviation = STDEV.S(LT1,LT2,LT3…LTi)

Average lead time and lead time standard deviation can be interesting, but don’t give us a full picture of lead time volatility on their own. Once we have these 2 metrics, we must compare them to one another in order to determine lead-time risk. To do that, we simply divide average lead time by lead-time standard deviation to get lead-time standard deviation percent (lead time risk).

The result quantifies lead time standard deviation as a percent of average lead time. Typically, safety stock is not justified anytime lead time standard deviation is 10 percent or less of average lead-time. Don’t forget, that lead-time deviation is the product of late and early receipt, but only late receipts create a risk of stock out.

That said, if the result is greater than 10 percent you should contact your supplier and request that they commit to reasonable lead-time. If they continue to be unreliable, issue a warning, before seeking out a more reliable partner.

Setting Supply Variation Safety-Stock
Using safety stock to shield against the effects of supply variation is not ideal. Ideally, we should be able to improve or replace erratic suppliers. Often, that’s much easier said than done. In the meantime, you must live with them, which means setting safety stock. There are a few ways to do exactly that.

1. Define target safety stock days as standard deviation of lead-time days. For example, say the average lead time is 14 days and the standard deviation is 3. In this case, 3 days of safety stock should be set as your safety stock. Understand that this may be excessive because standard deviation is based on early and late receipts. You’d want to keep an eye on this and adjust as necessary.
2. Quantify standard deviation only for late deliveries and define target safety stock accordingly. This mitigates the impact of early deliveries on analysis. However, it appears to assume that late delivery is acceptable. Remember, if you find yourself repeatedly dealing with late shipments, find a more reliable supplier.

With the above in mind, don’t forget to always communicate respectfully with all suppliers. Remember, good suppliers are partners to collaborate with not enemies to vanquish.

Respectful communication includes holding your organization to the same standards as your suppliers. Reliable suppliers should commit to standard lead times, but lead time us something that you, the OEM, must adhere to as well. If a supplier’s standard lead-time is 20 days, but you ask them to expedite 45% of all orders, it will appear as if they exhibit greater lead-time variation and might simultaneously suggest in an average lead-time that’s lower than the standard. This can cause a great deal of confusion during supplier negotiations.

3 Common Causes of Expedited Shipments
If you find your organization routinely expediting orders, that is something you will need to address internally. If such a situation, you are encouraged to conduct a 5-why or other root cause analysis. That said, it’s good to be cognizant of 3 common causes for needing to expedite shipments.

1. Cause #1 – Late order placements: Process orders faster. Increase the frequency that kanban signals are scanned and processed. If you’re only processing once a shift, start processing kanban orders twice a shift. If you use an MRP system instead of kanban, ensure that your MRP reports order exceptions and that they are promptly addressed.
2. Cause #2 – Incorrect MRP On-Hand Balance: Kanban is a lean replenishment process driven by actual on-hand inventory and consumption. Implement kanban to eliminate reliance on systems designed for planning rather than replenishment, while reducing on-hand inventory.
3. Cause #2 – High Demand: Be certain that demand variation isn’t the true cause of expedites. If demand is driving your need to expedite, you should acknowledge that fact and address it by adding demand safety stock.

Safety Stock for Supplier Quality
Even world-class supply chains suffer from the occasional quality issue. That said, quality errors should be rare. The quality-related reject rate should be less than 1 percent of orders (not item quantity) to avoid adding safety stock. Most business can manage reject rates below 1 percent without adding any safety stock.

If your reject rate is above 1 percent, you should determine the percentage of orders containing any rejects. It’s recommended to aim for an order reject rate of less than 2 percent but no more than 5 percent.

Is Safety Stock Really Needed to Address Quality?
Even world-class supply chains suffer from the occasional quality issue. That said, quality errors should be rare. The quality-related eject rate should be less than 1 percent of orders (not item quantity) to avoid adding safety stock. Most business can manage reject rates below 1 percent without adding any safety stock.

If your reject rate is above 1 percent, you should determine the percentage of orders containing any rejects. It’s recommended to aim for an order reject rate of less than 2 percent but no more than 5 percent.

Is Safety Stock Really Needed to Address Quality?
Simply because you receive orders with rejects, does not automatically determine the need for safety stock. If only a small part of an order is rejected and you sort out the bad items, then production may continue while you wait for replacements. In this case, safety stock will not be needed.

If entire orders are rejected regardless of the percent of the order that represents bad items, then you will need to set safety stock to address quality issues. In such cases, you have a few options in order of priority.

1. Your preferred option is for high-risk suppliers to commit to carrying additional on-hand inventory exclusively for replacing items rejected by your company. Keep in mind that this additional stock should be unavailable for purchase by any other business or consumer for any reason. While this can seem like a big request, a properly designed vendor managed inventory (VMI) program should automatically include such an exclusive inventory buffer.
2. If option #1 is unfeasible for any reason, you will need to set a specific amount of quality related safety stock. Minimal safety stock should be defined by multiplying the percent of rejected orders by the average number of days it takes the supplier to replace the rejected items. For example, if the percent of rejected order is 10% and the average days to receive replacement parts is 4 days then minimal safety stock days = 10% x 4 days = 0.4 days.
3. If setting the minimal amount of safety stock in option #2 is too risky, then it may make more sense to define quality safety stock in terms of the number of days it takes the supplier to replace the bad items. That is, if it takes the supplier four days to replace rejected items, then set quality safety stock to 4 days. Keep in mind, that this is a huge amount of additional safety stock. If your facility is this risk averse, then correcting the supplier issue should be a priority. Communicate your requirements to your supplier, understanding that the problem won’t be fixed overnight. Set a fair and logical time limit for improvement. That said, the issue should not drag on unresolved. Replace any supplier unable to commit to and achieve necessary quality improvements.

Supplier Target Safety-Stock Summary
Supply-related safety-stock should not be considered normal. Good suppliers should never necessitate safety stock precisely because quantity, quality and lead-time should be reliable. Unfortunately, many inventory managers see buffer inventory as a normal and appropriate solution to supply variation. The purpose of this 2-part series is to help managers grasp the causes of supply variation, and understand how best to mitigate its effects in the short-run and eliminate it’s occurrence in the long-run.

Think of your suppliers as collaborative partners. Remember to communicate expectations and measures to your suppliers. Good suppliers are able and willing to address any delivery and quality issues that may arise.

Any time additional safety stock is necessary to account for supply variation, it should be the supplier’s burden to bear. Clearly share your expectations and be prepared to replace the supplier if conditions aren’t met within a clearly defined and reasonable time frame. Carrying more safety stock should never be considered a long-term resolution.

That does it for part 2 of this 2-part series on supply variation safety-stock.