Build Less, Deliver More (the art of improving throughput)
Volume is the solution,
Except, well, it's not actually true, well not entirely. If you are in a high volume environment then possibly (but not actually), if you perhaps make 1 or 2 dedicated products then sure but if you are like the majority of organisations who have a range of products and low to medium volume then actually you are probably shooting yourself in the foot with this approach, killing your delivery performance actually increasing your lead time and quite possibly losing orders and money.
Lets think about it for a moment
Let's work through a process to try and explain this. Imagine you have a process that needs 5 steps through 5 different work centres or machines. They can only do 1 thing at a time so the longer they are doing 1 thing then the longer the queue is for everything else, it all has to wait it's turn. The larger the batch the longer it must wait. So that's a big part of your lead time issue right there isn't it?
What about Throughput rate? i.e. how fast things go through your organisation. You want that as fast as possible, so you turn your raw materials into sellable items as fast as possible and ideally get someone else to buy it before you have to pay for the raw materials. Well if your batch size is too high and your waiting time for the process is also high then the chances of doing this are really small and guess what, you reduce throughput (which increases lead times & increases costs since you have to fund the WIP)
Now because of this issue what normally happens is that you cleverly decide to increase your stock holding so you can respond to clients quicker. So Stock holding & financing costs go up, you also have to build larger batches to build up the stock so your WIP goes up, lead time is up and throughput is down…. Wait isn't that the opposite of what you wanted?
Here is a visual, so if you keep the numbers simple say you only build 5 widgets at a time, each widget takes an 12mins at each operation and you batch in groups of 5 so it's an hour per operation (since this is a mystical process with perfectly balanced operations)
Here are the basic results you get from this system
|Time to get 1st batch out||5 hrs|
|2nd & Subsequent Batches||1hr|
|Qty built in 8hrs||20|
|Value of WIP||$25000|
This is the WIP profile so you can see how the 20 widgets per shift occurs
So it takes you 5hrs to get anything out the system then you get 5 every hour after that, so if you run say 8hr shifts you would get 20 widgets in total (5 after 5hrs, then another 5 for hr 6,7 & 8)
Now Reduce Your Batch Size
What happens when you reduce the batch size to 1 (we know you may never do this but it's an illustration)
Here are the basic results you get from this system
|Time to get 1st batch out||1 hr|
|2nd & Subsequent Batches||12 mins|
|Qty built in 8hrs||36|
|Value of WIP||$5000|
This is the WIP profile so you can see how the 36 widgets per shift occurs
Surely this isn't right?.. think about it, your 1st widget gets through the system in 1hr, then every 12 minutes another pops out, then another & another. The 2nd operation isn't waiting an hour to start, it's waiting 12 minutes, the 3rd operation instead of
Now obviously the numbers will vary a little over the batches you do and the size of them but think about it, if you needed to supply 40 widgets would you rather tie up your line for 9hrs or 16hrs? what could you build in that free 5hrs?
What do I mean by free? Well it's not just that the equipment is available but the people as well, it's that it's actually not costing you any more to use that 5hrs than you had planned (and probably quoted) for the last job. So if you rejigged the process and costed the job against the smaller batch size the time for the line is actually less.
So lets assume the line costs you $1000/hr, material costs you $10, you sell the widgets at a 30% mark-up so each widget sells for $1313. Remember you are using 7hrs less of line time than the original method so that's $7000 (19%) additional profit or you can lower the price slightly and undercut the competition and deliver faster.
One thing to remember however is that the time to build the batch size should not be lower than the setup time or you risk making the setup time the constraint. If you want to reduce the batch size lower you need to reduce the setup times.
These are basic concepts of the Theory of Constraints (TOC) which is about improving flow & tightly managing your bottlenecks. In this scenario, we haven't created any bottlenecks but we will explore this in later discussions.
What about LEAN? Well if you think about the Lean Wastes (I like the DOWNTIME list of wastes) i.e. things that don't add value, in fact, cost you money these are:
- Defects: Smaller batch sizing allows you to see defects quicker, correcting them quicker and so reducing the number of defects and saving cost.
- Overproduction – by reducing the batch size you can better match the qty you need to match sales & reduce inventory on hand since you have higher throughput
- Waiting – smaller batch sizes as we have shown flow through the LEAN / TOC style environment faster, so work centres spend less time idle
- Not Utilising Talent – by using smaller batch sizes operators are engaged sooner, by including them in the challenge of reducing set up times they become even more involved in driving down batch sizes and of course eliminating defects.
- Transportation – smaller batch size doesn't directly impact transportation in a negative way, in fact, it would encourage you to reduce the area used!
- Inventory Excess – Inventories can be reduced as you have faster throughput and can respond quickly to requirements freeing up cash tied up in inventory as well as space.
- Motion Waste – No direct effect
- Excess (over) Processing i.e. building a Rolls Royce when the client wants to buy a Toyota. Again the smaller batch size doesn't directly impact here
By sensibly reducing your batch size you can actually reach your goal, you can dramatically increase your throughput and reduce costs as we've shown above. It doesn't matter if you take a Theory of Constraints (TOC) viewpoint on a LEAN viewpoint the process is essentially the same. When you want to improve your throughput, delivery performance, increase your market responsiveness and reduce costs the answer really is to Build Less & Deliver more.
Get In Touch
If you need any support in understanding or improving your bottleneck performance then click here to make an appointment and find out how we can help you Make Things, Better
You can also call John on 0211649739 to set up a meeting
© Many Caps Consulting Ltd | All Rights Reserved