ATO and factory load Part1
Twitter
Facebook
Delicious
Digg
Stumble
Reddit
Lean manufacturing essentially started as a means to react to changes in demand in a way that would not result in losses to you and to your suppliers. In one sense, you do this by moving the ‘inventory holding centers’ upstream — i.e. instead of you holding inventories for finished goods, your suppliers, or their suppliers, or theirs, might keep inventory. The rest of the chain just works in a real-time manner to incoming demand.
Parts suppliers — holding stocks of, says, buttons, would be able to assume the risk of holding stocks because they can always sell buttons to someone else if demand for your product changes.
Assemble-to-Order (as opposed to build-to-stock) is one manufacturing process that relies on an agile supply-chain to be successful. ATO can be used to achieve zero-inventory, but most prominently it will help you minimize your working inventories.
After the link I very briefly introduce the benefits of ATO. Without going into theory of how to align your supply chain for ATO manufacturing. I was to start analysing this within the context of Pakistan.
If you are assembling the final product only after a customer places an order, you dont have to invest in large build-up of inventory up front. If the manufacturing and supply engine is fast enough (read: agile), you are also able to get out of ‘factory-driven’ limitations and move into a ‘customer-pull’ model.
Dell, the #2 supply-chain in the world (after Wal-Mart) does this very well. Customers configure their system online and it is assembled, packaged, and delivered from Malaysia to the USA within 10 days.
Depending on the customizations in your product, and the volume of your ‘predictable’ demand, you can even have the customer walk into your door, order something, and have you give it to him immediately. This is because you kept a minimal inventory with all of the anticipated configurations. The ‘order’ that the customer placed is still used in the ‘assemble-to-order’ engine, but the finished product in fact will be sold to someone else later as it replenishes the inventory.
Ofcourse with the above you had a limited predictable number of configurations. With greater customization (cars, PCs etc.) you will have to give your customers a lead-time for the manufacturing. E.g. you might be able to deliver the final product within a week or so. This promise is quite difficult to make — I hope we see this in later posts — so it is actually remarkable that Dell gives you a final ship-date on its website in real-time.
Depending on the lead-time, the inventory levels you need, characteristics of your demand, etc. many important models can be made of your business, such as: Who really are your immediate customers? What sales-channels will you use — retail? Partnerships? E-Commerce only? What will your distribution network look like in order to guarantee your lead times? How do you react to demand fluctuations and new market opportunities?
Note: this shouldn’t be confused with lag capacity — lag capacity building applies to the project as a whole : everyone in the supply chain, from you, to your suppliers, to their suppliers, will not do any preparation before a clearly identified ‘project’ has been created.
With lean manufacturing, you know that, say, the ‘project’ invovles building 40,000 units of your product, based on which everyone in the supply-chain can build their capacities. Based on this preparation, everyone works in real-time to build each individual unit against each individual order placed. The advantage here is in being able to react to increased or decreased demand against your forecasts.
Facebook comments: