Throughput Accounting: Financial Analysis (#46)
/In this podcast episode, we cover a variety of scenarios that can impact the amount of throughput generated by a bottleneck operation. Key points made are noted below.
In this episode, I’m going to run through a bunch of financial analysis scenarios, so you can see how throughput applies to it.
The Base Case Scenario
First, let’s set up a base case scenario. We have orders for two products. Product A has a total throughput of $10 and requires 5 minutes of time at the bottleneck operation per unit, so it generates $2 per minute of bottleneck time. That’s the throughput of $10, divided by 5 minutes. There’s a customer order of 500 units for Product A, so that requires 2,500 minutes of bottleneck time.
Next, Product B has a total throughput of $20, and requires 4 minutes of time at the bottleneck operation per unit, so it generates $5 per minute of bottleneck time. There’s also a customer order of 500 units for Product B, and this requires 2,000 minutes of bottleneck time.
Scenario #1
So, here’s our first analysis problem. The bottleneck only has 4,000 minutes of time available per week, but we have 4,500 minutes of required production time between the two products. So, what do we not produce? The answer is to always produce the maximum amount of any product that generates the largest amount of throughput per minute, which in this case is Product B. It generates $5 per minute, versus $2 per minute for Product A. Therefore, we produce all of the order for Product B, and we intentionally fall short of production on Product A, which maximizes profits.
So, let’s extend that example. The sales manager, bless his heart, gets a huge order for Product B that can tie up all of our production capability, but only if we accept a price drop that reduces throughput per minute to $4 from the original $5. Is this a good deal, or do we fire the sales manager?
The solution is actually pretty simple. Under the base case, the total throughput we earned was 2,000 minutes of Product B and 2,000 minutes of Product A, which totaled $14,000. Under the new scenario, anything better than $14,000 is a good deal. Since we’ll now be running 4,000 minutes at $4 of throughput per minute, the new total throughput is $16,000. So, we give the sales manager a cookie.
Scenario #2
Let’s try another situation. We receive a proposal from the production manager, where he wants to pre-process some of Product A before it reaches the bottleneck operation, so that it requires less time at the bottleneck – in fact, it only requires 2 minutes per unit, instead of the old 5 minutes. This increases its throughput to $5 per minute of bottleneck time. With the faster processing speed, we can now run all of both orders, for a grand total throughput of $15,000. We get that number from $10 of throughput for Product A, times an order size of 500 units, and $20 of throughput for Product B, times an order of the same size.
So we’ve established that our throughput will increase by $1,000 per week if we implement the change. The real question is, what investment and added expense do we incur in exchange for the $1,000 throughput increase? This becomes a management decision that depends on the circumstances, but – for example, what if the investment was a one-time outlay of $5,000? Then it would pay for itself in five weeks, and I think most managers would accept that proposal in a few seconds.
Scenario #3
Let’s try another example. How about if there’s a quality problem with Product B, so that an average of 100 units are scrapped per week, after they’ve been processed through the bottleneck? Well, each unit earned $20 of throughput, so we’re losing $2,000 every week because of the quality problem. The question is, should we install a quality review station directly in front of the bottleneck whose sole job is to review Product B parts?
The answer is - yes, as long as that quality workstation costs less than $2,000 per week.
Scenario #4
OK, let’s do another example. What if the employees who run the bottleneck operation take some time off each day for breaks, so that the work station has total weekly downtime of 200 minutes?
In this case, Product A has less throughput per minute, at $2 per minute, so it will always be scheduled to go after Product B, and therefore if there is downtime, fewer units of Product A will be completed. Since we’d be losing $2 of throughput per minute times 200 minutes, there’s a potential profit loss of $400 per week. So, if we can get a part-timer who costs less than $400 per week to fill in during those breaks, then we’ll make more money.
The Product Cancellation Decision
And then we have one of my favorites, which is the product cancellation decision. The cost accountant applies $11 of overhead to each unit of Product A, which now appears to give it throughput of -$1, and makes a recommendation that we therefore cancel the product. But, if we did that, the bottleneck would no longer have enough work to do, and we’d be walking away from $5,000 of throughput – just to repeat, our base case was $10 of throughput per unit, times an order for 500 units, which is $5,000.
The solution is to ignore the overhead application, we do not cancel the product, and we keep on producing Product A. This gives us more profit than if we cancelled Product A.
But there is a situation where we do cancel Product A, which is when the engineering department creates a new product. In this case, if the throughput of the new product is more than the $2 per minute of bottleneck time that we had with Product A, then the company as a whole will make more money if we dump Product A in favor of the new product.
When There is No Bottleneck
And finally, what if the bottleneck has enough capacity to produce every possible order? In this case, there is no production bottleneck. Instead, the bottleneck may very well be located in the sales department instead. Or, it may be located outside of the company entirely, in which case the company might actually make more money by dropping its price in order to bring in more order volume. As long as the new price is not too low, the company could use its extra production capacity to crank out more throughput dollars than it had before.