One thing that you might notice on your schedule is a negative TDD value. This will occur when the due date has not arrived yet. It should be a positive value if the due date is today or greater. But in order to get that there are a few tweaks I would recommend.
Let’s tackle the last one I mentioned. If you use a date difference function from Microsoft and the due date is today and you compare that to the Now() or Date function you will get zero. I personally don’t like zero for a TDD because it immediately negates the other two values your are multiplying it by. So wrap your date diff function in an if statement so that if the value is zero for the days value it returns a 1 so you still get a TD calculation when it’s due today.
Now let’s discuss why a negative TDD value should be tweaked. Most companies have an additional limitation on resources, meaning they have to batch at some level, and this is often referred to as the MOQ or minimum order quantity. It’s the minimum order size that they can run. This added constraint is often compounded when like products or services are ran back to back to minimize change overs to an extent. This the result is batching. TOC is opposed to batching as far as it can be realistically controlled. I will discuss this in more detail later.
Side note: people think that small batches and change overs are inefficient. But if they are on non bottleneck machines then they are not significant. If it is a bottleneck machine or process you need to have a pit crew mentality for the bottle neck only! Obviously I am not saying take as much time as you want on the other change overs.
So why don’t we want negative TDD again. Here is how we make the connection. If you separate or group your schedule by processes that line can change over to and sum the TDD for each of those processes on the constraint schedule you now know what processes should be first and second and so forth. It tells you what to set up to and what to go to next. You can even quickly calculate when to change over to the next type or process by calculating how many jobs ran under the top TDD process it takes to lower that processes total TDD below the second highest TDD process. And if you are within a job or two that’s good enough.
Side note: one major struggle companies have with their schedules is that they always have an alternative schedule, the hot list or the expedite list. Call it what you want but they are hot jobs that need inserted into the schedule regardless. Someone promised the customer and it’s everyone’s job to jump through hoops to get it done. The cool part about this TOC schedule is that it allows hot job inserts at calculated locations or buffers in the schedule. You can simply put it in the schedule at the beginning or end of the process it fits into. And with smaller batch sizes and more frequent change overs on non-constraints that shouldn’t be a problem. Plus if you do it right after a few of the next posts and some time implementing you shouldn’t need to expedite any jobs any more. More to come on this soon.
Back to why not to have a negative TDD. If you add the features in your schedule that I mentioned above a negative value will subtract from the sum value of the TDD for each process. And if you are like most companies that typically have a smaller back log when compared to scheduled jobs your total sum value will be significantly reduced to the point you can’t use this value to arrange the groupings of your orders by process and the priority of those processes.
The solution is to take the inverse or reciprocal of the absolute value of the date diff function (1/abs(datediffvalue)). This way the farther out the due date the closer the Days value of TDD approaches zero but never is zero. And as it approaches the due date it will approach the value of one.