I have been asked this before and I think it’s voodoo math, but tell me what you think.
Let’s say:
Today is February 1
Supplier says it will take 3 months to get part XYZ
We place a PO today (Feb 1) asking for the parts to arrive on September 1
Parts arrive August 1
According to my users, the “real” lead time is 6 months (Aug 1 minus Feb 1).
But really it didn’t take the supplier 6 months to source the parts. We simply planned ahead. How long did it actually take them to make the parts? I don’t think you can tell. (And really, they may have used the wiggle room we gave them in order to squeeze in some other orders.)
I feel like I can’t ever know the “real lead time” in a facility that plans ahead. Right? I mean, I could, of course, give them the report that does the math the way they want. But I don’t like putting my name on bogus data.
(This math does work for our service division, though, because they don’t plan ahead - they are reactive.)
Also, in this example, the supplier messed up, because we didn’t want the parts that early. But that’s not part of the discussion, I guess.
I came to the same conclusion but ended up making the report anyway. I think it’s important to consider that you can’t accurately do this with blanket type POs. But for more discrete POs, you could argue that you should not place a PO for further out than you need and therefore your PO should be placed at the appropriate time (supplier lead time) before the parts are due - based on a suggestion. So you can compare their lead time against the time between PO Order Date and the receipt date. To try to disqualify POs that would not fit neatly into this evaluation, I also totaled the number of releases as well as the number of POs and you get a weighted average of how many releases there were per PO for any given part. If it’s exactly 1, then it should fit. The higher the number, the less valuable the report is for that part. At the very least, going through this exercise can help to shed light on where the actual issue might be and could give you insights on where it might make more sense to build a report.
I’ve never worked in purchasing, though I have been around it A LOT. So to be fair, I don’t know “the right way” to buy things according to the Society of Purchasing or whatever trade groups exist.
But I would think that if you had this nifty MRP system that lets you plan months in advance, then you’d use that to give your suppliers extra heads up to get you the parts in, in time.
It definitely seems like we do not do that here; I hear a whole lot more of panicking that a supplier is late and we need these parts tomorrow, etc.
So I guess my idea is not the norm? That you do in fact wait till the last minute to place the PO? (Last minute, meaning per the suggestions, as you outlined.)
I know I’ve announced on this place before - but I work for one company and I’ve only ever worked for this one company so my exposure to all the different ways people do things is minimal.
I also am not a member. lol. But I would think they have a code they follow as far as long term planning strategy similar to the sales side where you try to feel out your customers for the next year or so and then you can enter forecast. Forecast (in Epicor) drives the long-lead parts that you can act on before firm demand. I assume supply chain does the same thing in reverse. But you don’t actually have firmed POs until you need to.
This is one of those things that you’ll need to look at over time and see if the data you get is LESS right, MORE right, or pretty much the SAME right that whatever other source you’re using. Any time we try to predict what will happen tomorrow we’re liable to be wrong.
I’m in the camp of “the more ways I can try to predict something, the more likely I am to get closer to it”.
One more thought on that topic. That’s what our CEO was asserting. And he wanted to understand why. He wanted to know if it was our own people not reacting to suggestions in a timely fashion, our suppliers not performing, or our ERP system that was lacking. So we ran through a series of reports that ferretted that out.
I think you’re overthinking this @JasonMcD. It doesn’t matter (to you, the customer) how long is actually takes the supplier to do something. What matters is how long contractually, you agree on from when you place the order, and they agree to deliver. The amount of “work time” in any product is usually only hours. The real amount of lead time is how much other work is in line before yours, and that’s highly variable. Suppliers say “Six to Eight Weeks” in order to give them leeway to manage the schedule. Whether or not you allow the vendor to deliver early or not is up to how your terms are dictated.
This is actually right, you just don’t have anything in place that says they aren’t allowed to deliver early. Having your system put in more lead time than the supplier requires is a nice luxury to have because if they can’t meet their own lead time, it doesn’t put you over a barrel.
We have sales orders out for many months in advance
We schedule them as we deem fit
We place POs in order to get the parts in time to build in order to fulfill to the schedule
We aren’t “just in time” per se, but we can’t have parts getting here just whenever, nor do we have endless storage space or limitless working capital that we would want the stuff too early
We are not alone in that, right?
Well, exactly. So why calculate anything? Fundamentally it’s up to them, no?
So how do you and others place POs? I’m truly curious.
Using that same fictional supplier, do you send out the PO 6-8 weeks out? Or 10 weeks out? Is there ever a time you order 20 weeks out, and if so, do you tell them when you need it by, or just not tell them and see how long it takes?
This is NOT a “technical” answer… but it’s the best answer I’ve found.
When Buyers have good relationships with their Suppliers, they get better information. This is a human issue. Our sales people want their customers to like them so they buy from us… well, to the supplier we are the customer. The better we treat them the better they’ll treat us.
You need to negotiate with the supplier. Large automotive companies have penalties for being late (obviously) but ALSO penalties for being early. If you can’t have stuff early, you need to line that out in your PO and have the supplier acknowledge that. And if it’s really a big deal, put penalties on it.
You need to tell them what you need, not try to guess when you need to send the PO’s to try and get stuff at the right time.
Tell “them” that you can get min, max, and average delivery times… but in your mind those numbers may not have direct bearing on an “actual” lead time value. In a perfect world, this is more a question for the Buyers… but if they don’t have the information, all you can do is what you’re doing.
