ATO and factory load Part2
Twitter
Facebook
Delicious
Digg
Stumble
Reddit
We discussed ATO basics a few days ago.
In order to implement ATO, you will naturally go to your factories and ask them to move away from requiring bulk orders up-front, and move towards a long-standing agreement to assemble products according to your customers’ orders.
The biggest concern you will hear from the factories is that of load. “What if you give us 15 orders one week, and 5 the other. What happens then?”
The concern is a very valid one, and infact will be core to any solution. Any solution will be bounded most by the load on the factory, and end-customer price point, among other things.
As their customer, you’re asking the factory for a specific manufacturing cost per unit so you can meet your target price point to your customers.
The factory can only give you that cost per unit if it can plan its resources with 95-100% capacity load. It’s overheads will simply be divided across the number of units manufactured. In this way, the factory can also plan material and parts purchases in a way that they are used up. Remember, you will end up paying the interest on short-term loans your manufacturing partners take to build up material supply.
If demand fluctuates on your customer end and the factory load becomes 75%, not only do the parts stocked at the factory start aging, the cost per unit given to you increases.
If you own your factories, then you may want to pick your poison — is it worse off to bear the extra cost of running at 75%, or is it worse to have 5000 unsold extra units (that you built to keep load to the max) when demand has died?
With outsourced manufacturing, you dont have enough influence on your partners to ask them to work at sub-capacity. So ATO designs must ensure that factories can realize maximum economies. Fortunately, this CAN happen easily in Pakistan with two types of demand:
1- If your demand is perfectly stable, then the factory can rest assured that the parts and supplies will turnover as expected, and also that the factory size and resources can be adjusted for optimal load based on the forecast.
2- If your demand does fluctuate, but that *fluctuation is bounded*, then you can tell the factory that you will atleast give him the lower edge of the demand, with a possible upsell up to the upper edge. The factory will allocate its machines and resources based on the lower-edge (to ensure 100% capacity) but keep stock of supplies based on the upper-edge. In case demand does infact rise, the factory would just have to schedule overtime, which is much easier to do on a short notice than purchasing additional supply.
Another area depends on the margins on the product. Sometimes you may want to let the cost-per-unit change based on the changing load on the factory, and let your margin act as a buffer to adjust the variations. This will let you keep the same price-point, but will give you variable margins.
If your demand fluctuates too much, or without predictable bounds, you would end up with inventory or back-orders. You want to keep ensuring that the factory remains at 100%, so whenever demand goes down your own inventories will rise. In such cases you would want to immediately jump on building demand through promotions.
There are hybrid approaches and clever tricks you can play to adjust the system as well. What do you think are good, smarter hybrid models that can work?
Facebook comments: