Understanding Cycle Time - Why the time of the other activities are left out

[zaffresky]

Hi everyone. I am a new member of this forum and currently learning industrial engineering concepts on my own. I look up the definitions online but at times I couldn't get it. For instance, a lot of online resources mention that cycle time is determined by the slowest activity or task (aka bottleneck) in the process. I do understand that cycle time will be a function of bottleneck but why the time of the other activities that constitute the process are left out. I believe the cycle time should be the sum of all process activities?

Can anyone please help me understand that?

 

[normzone]

Welcome to the forum, [zaffresky].

I don't know that I can help you, but I also have some interest/questions in that area, so I'll contribute to the confusion in that manner, and somebody will be along shortly who knows what they are talking about. Perhaps from your studies you can help me out.

I'm going to guess that there are different methods of referring to cycle time, and the method you reference (including the time of ALL process activities) has it's own word, unknown to both you and I.

My curiosity regards our inhouse statistics for why we ship late - one of our causes is referred to as "labor". It's the biggest one, since we're low volume / high mix, and we have to staff up and down depending, just as I'm certain everybody else does to some extent. But since it's the biggest one, I'll wager there are other factors buried in there - friction from resetting the line (moving components in and out for differing builds) etc.

Can you shed any light on this? Thanks, and apologies for piggybacking on your thread.

 

[Candi1024]

Cycle time is how long it takes for a single process to be completed. Another way to explain it is how long is the time between outputting a single completed piece? If the cycle time was 30 min, (and you made 1 piece at a time) the production rate would be 2 per hour.


Lead time is how long it takes for one single product to travel through the production flow. That would be the sum of all the process times.

 

[Michael_M]

I will use a modified example of what we do for cycle time (assume each step is run after the previous step is completed):
Lets say 100 parts


Step...Process............Time each.........total time
10......Lathe turn........45 sec..............4500 seconds or 75 minutes
20......Lathe finish......30 sec..............3000 seconds or 50 minutes
30......Mill Finish........120 seconds.....12000 seconds or 200 minutes

With this pattern, we have 19,500 seconds or 325 minutes to run all 100 parts. Now lets combine step 10 and 20 into a single operation:

Step...Process..............Time each.........total time
10......Lathe Complete..45 sec..............4500 seconds or 75 minutes
20......Mill Finish..........120 seconds.....12000 seconds or 200 minutes

Since step 10 is running what was step 10 and 20 at the same time, we use the highest time (45 seconds in this case) since the 2nd machine is running at the same time as the first. This reduces the cycle time to 16,500 seconds or 275 minutes.

Note: This is a very basic version. It does not take into account setup times or other steps required.

 

[normzone]

" Lead time is how long it takes for one single product to travel through the production flow. That would be the sum of all the process times. "

I can readily agree with the first part of that statement, that's how we use it.

But the second part doesn't appear to take into account concurrent processes.

EDIT: Ah, [Michael M] got in there while I was posting.

 

[Bev D]

Candi1024 is close...Normzone is correct in the sense that many people just cant' resist changing the definition of common words.

Cycle time is an internal manufacturing time. It is the time it takes a single part to complete the process, either individual process steps or the entire process including parallel operations. It includes queue time, idle time, downtime, actual touch time, whatever. It is the time from start to finish. For a single process, the clock starts when the part competes the previous step and it stops when the part completes the step for which you are calculating the cycle time. When dealing with total cycle time you must decide on the actual start time. Is it when the parts hit the floor or when the order is released? things like that. You also need to clearly identify the 'completion time'. is it when the part hits a shippable location? or is once all manufacturing steps are complete? But you need to be honest with yourself if you want to make real improvements to cycle time or use it for accurate planning purposes.

Lead time is a Supplier - Customer term and includes all of the time from ORDER to DELIVERY

I suspect that the source that identifies 'cycle time' as the bottleneck is trying to say that the bottleneck will control cycle time (which is true) or they are trying to make up their own definition of cycle time...

 

[Bev D]

My curiosity regards our inhouse statistics for why we ship late - one of our causes is referred to as "labor". It's the biggest one, since we're low volume / high mix, and we have to staff up and down depending, just as I'm certain everybody else does to some extent. But since it's the biggest one, I'll wager there are other factors buried in there - friction from resetting the line (moving components in and out for differing builds) etc.

Can you shed any light on this? Thanks, and apologies for piggybacking on your thread.

Perhaps this question belongs in it's own thread?

in the meantime, let me just say that in my experience this assigning of 'causes' to late shipments (or anything else for that matter) is usually less about trying to understand the causal mechanisms of something and more about assigning blame...they are not grounded in facts and are such over aggregated 'categories' as to make them useless in driving improvements...

 

[normzone]

[Bev D], you're correct that a discussion of this issue belongs in it's own thread.

I agree that such categorizations of causes are often used to play the blame game, but in this case my goal is to determine what our actual process capability is, and look for possible process improvements.

In collecting and analyzing data I'm also hoping to be the benefactor of a phenomena I've often seen - you begin collecting and analyzing data and the problem goes away as soon as you begin looking at it regularly. Can't hurt to hope

 

[Bev D]

[Bev D], you're correct that a discussion of this issue belongs in it's own thread.

I agree that such categorizations of causes are often used to play the blame game, but in this case my goal is to determine what our actual process capability is, and look for possible process improvements.

In collecting and analyzing data I'm also hoping to be the benefactor of a phenomena I've often seen - you begin collecting and analyzing data and the problem goes away as soon as you begin looking at it regularly. Can't hurt to hope

First hope isn't a business strategy. And the study that you are referring to was misinterpreted, mis-reported and mythologically repeated. The Hawthorne Effect was discredited years ago, see: "Was There Really a Hawthorne Effect at the Hawthorne Plant? An Analysis of the
Original Illumination Experiments" by Steven D. Levitt and John A. List
The end result is that any improvement you may see are 1. coincidental and not causal and 2 temporary at best.

Second I would recommend really studying what causes you to miss deliveries. what are you collecting on? how are you investigating the causes? remember shallow knowledge leads to shallow solutions and/or analysis paralysis.

You can compare what is different between orders that are on time and orders that are late. What are the common factors? be specific, don't categorize. do the misses have longer cycle times? were there unique assignable causes such as snow storms associated with the misses but not the on time orders, are certain products more likely to be late than others? etc. Often - but not always - the missed deliveries have longer cycle times. If this is the case you can look for where cycle times are the longest, look for bottlenecks, look for waste to be removed, look for defects to reduce or eliminate. But be specific; the cause lies in the details. Study the Toyota Production System...

 

[Miner]

Here is a good illustration from Harvard Business School.