I want to live in I.E. world

Omega man

Well-Known Member
Here's a thought--if you know that this COD customers is slow, why not give them a call with the COD and PO information so that the checks will (hopefully) be ready when you make the delivery? I have this one stop which gets multiple CODs and I will write down the COD information, hand it to her, complete the rest of the stops on that part of the street, and then go back to her and she will have all of the checks ready. Takes less than 30 seconds and saves me 5 minutes.

Yes, we do get paid by the hour, but there is nothing wrong with trying to use that time more efficiently.


Can't do. Simply, because this is not part of the methods.
 

TechGrrl

Space Cadet
There are no allowances made for everything you mention. That does not mean that your last statement is true.

Time studies are worked up initially based on "expected conditions". Not perfect conditions. The "expected conditions do not include the ones you mention, but.... This is an important but...

After the base times are determined, an additional allowance is added. I don't know what its called today, but when I was taught, it was called PDF.

Nearly 30 years ago I went to school at UPS to learn how to do original time studies. This is different than the ones I.E.'s do when they study you. A select few people were sent to school and we got a coveted "blue card" that enabled us to work up and original analysis.

The last step of working up a study was to add in PDF. It used to be 15%. It was meant to account for many of the things not accounted for. PDF stands for Personal, Delay, and Fatigue.

So, its not true that perfect conditions are expected. Its not true that no delays are accounted for.

Also, as I have said before, work measurement is NOT meant to measure an individual driver for an individual day. PDF may be too little for you, and too much for a different driver.

P-Man

P-Man, I had one of those little blue cards, too. And I agree with everything you say, regarding the THEORY of work measurement as you and I were taught it.

However, the sense I get, is that in the wonderful world of today, the IE function has lost sight of that messy theory that life is a bell-shaped curve, and that actual performance will fluctuate around the allowance + or - 5%. (Don't forget the fact that we only studied to be 95% accurate 95% of the time!)

So, that means, given a stable route, and a skilled driver, the actual will fluctuate, a little above, a little below, from day to day, but should average out to be on plan most of the time.

If a particular route is consistently too high on off performance, or consistently too low, there is something wrong with the measurement. (again, assuming the driver is skilled and familiar with the area)

But I get the impression that the cost of doing actual, real live time studies is something IE doesn't want to incur any more, and they do a lot of it off the DIAD. And don't get me started on using college kids to do the studies. We did that back in the 80's, and it was a disaster: the drivers led them around Robin Hood's barn and left them wondering where they were.
 

UnconTROLLed

perfection
I agree that timestudys shouldn't be a priority but it is humorous to see the OR's and 90% of the drivers aren't doing their routes fast enough??
Really?? Almost every run is paid over??

This isn't a case of the driver not doing his/her job properly or stealing time. It is a simple flaw in the system. One, might I add, that never seems to be addressed.

At least tease us a bit and fix some of the allowances so they are actual and not fabricated. Not one of the 3 mall runs in our center can be scratch. It's the numbers, not the drivers.

I'm fairly new to this so bear with me - if 90% of the drivers WERE running scratch and even underallowed according to the time studies, that would like lead to a complacent attitude amongst drivers - in the eyes of higher ups - would it not?
 

Omega man

Well-Known Member
If UPS' work measurement was truly fair (based on performance averages) and accurate the following should apply:


1. The MEAN value for the entire sample of drivers should be 0.
The MEAN is the average (adding together all of our results and then dividing it by the sum of the total).
That is: If you total up the performance results of every driver (+and-) and divide by the total number of drivers it should be 0 (scratch).

2. The MEDIAN value should be very close to the MEAN.
The MEDIAN is the 'middle value' in your sample.
For example: if 5 drivers ran -.30under, .30over, .70over, .80over, .90over then the MEDIAN is: .70over


3. The MODE value should also be close to 0(scratch).
The MODE in a list of numbers refers to the value that occurs most frequently.
That is: There should be more drivers running scratch than anything else.


As everyone knows, most of us are running way over-plan. So, how can this be a fair work measurement? It is NOT!
 

pretzel_man

Well-Known Member
P-Man, I had one of those little blue cards, too. And I agree with everything you say, regarding the THEORY of work measurement as you and I were taught it.

However, the sense I get, is that in the wonderful world of today, the IE function has lost sight of that messy theory that life is a bell-shaped curve, and that actual performance will fluctuate around the allowance + or - 5%. (Don't forget the fact that we only studied to be 95% accurate 95% of the time!)

So, that means, given a stable route, and a skilled driver, the actual will fluctuate, a little above, a little below, from day to day, but should average out to be on plan most of the time.

If a particular route is consistently too high on off performance, or consistently too low, there is something wrong with the measurement. (again, assuming the driver is skilled and familiar with the area)

But I get the impression that the cost of doing actual, real live time studies is something IE doesn't want to incur any more, and they do a lot of it off the DIAD. And don't get me started on using college kids to do the studies. We did that back in the 80's, and it was a disaster: the drivers led them around Robin Hood's barn and left them wondering where they were.

Tech,

I've been out of I.E. for about 10 years. I do know they stopped doing studies for a while, but then started it up again. I don't know how much is observed vs. using data. I do know that they are NOT doing automated time studies (obserless). I think its a combination of the two.

There is one thing that needs to be added to the paragraph above that I bolded.

You have to add to that that the route has a proper job setup. The time study values assume not only proper methods, but that the route is properly loaded.

From what I've seen, this is the single most important area to address. The average driver is over 20 minutes overallowed in select.

Does that mean that the measurement is wrong or that the job setup is not right. As you know, this has been a long debated topic in I.E. If we adjust the allowance, the issue never appears as somethingn to work on.

In the case of select (which is the single largest overallowed area), it is not the driver's fault nearly all the time. The cause is the preload.

Also, interestingly enough, the overallowed in select got worse after PAS / EDD.

Because of these things, I do not think it follows that "If a particular route is consistently too high on off performance, or consistently too low, there is something wrong with the measurement. (again, assuming the driver is skilled and familiar with the area)"

If you add proper job setup to the mix, I would then agree. (I think I'm going to look for my blue card, just for nostalgia.

P-Man
 

NHDRVR

Well-Known Member
I'm fairly new to this so bear with me - if 90% of the drivers WERE running scratch and even underallowed according to the time studies, that would like lead to a complacent attitude amongst drivers - in the eyes of higher ups - would it not?

Good Point...

In other words....It's never enough....
 

pretzel_man

Well-Known Member
If UPS' work measurement was truly fair (based on performance averages) and accurate the following should apply:


1. The MEAN value for the entire sample of drivers should be 0.
The MEAN is the average (adding together all of our results and then dividing it by the sum of the total).
That is: If you total up the performance results of every driver (+and-) and divide by the total number of drivers it should be 0 (scratch).

2. The MEDIAN value should be very close to the MEAN.
The MEDIAN is the 'middle value' in your sample.
For example: if 5 drivers ran -.30under, .30over, .70over, .80over, .90over then the MEDIAN is: .70over


3. The MODE value should also be close to 0(scratch).
The MODE in a list of numbers refers to the value that occurs most frequently.
That is: There should be more drivers running scratch than anything else.


As everyone knows, most of us are running way over-plan. So, how can this be a fair work measurement? It is NOT!

See my previous response to TechGirl.

The mean, mode, and median will not all equal out if the job setup is not correct. That is the single biggest issue I have seen.

For example, a driver that sorts his load will be overallowed for that time. There will never be an offsetting value there. That is not part of the job setup.

P-Man
 

soberups

Pees in the brown Koolaid
I understand your frustration with the measurement process but I don't think the timestudy is a priority. The basic concept is if your a buck over today can we do something other then another timestudy to get you .75 tommorrow.

Unfortunately, that "something" you can do usually consists of endless OJS rides, meetings with shop stewards, warning letters, constant harassment, over 9.5 grievances due to excessive dispatch, etc.etc.---none of which will ever solve the underlying problem of the flawed measurement.

The energy we waste fighting and bickering with one another over these unrealistic expectations could be better spent cooperating in our shared struggle to gain volume from FedEx.
 

soberups

Pees in the brown Koolaid
See my previous response to TechGirl.

The mean, mode, and median will not all equal out if the job setup is not correct. That is the single biggest issue I have seen.

For example, a driver that sorts his load will be overallowed for that time. There will never be an offsetting value there. That is not part of the job setup.

P-Man

You cannot blow a package car out and then expect the driver to deliver the route without sorting.

The fact that it is not "part of the job setup" in no way changes the fact that it is necessary to do.

Maybe the job setup isnt correct, but it since it will never be corrected, the allowance needs to be based upon the conditions that exist in the real world rather than some utopian pipe dream where every load is perfect, every car is new and big enough, and every condition is ideal.
 

pretzel_man

Well-Known Member
You cannot blow a package car out and then expect the driver to deliver the route without sorting.

The fact that it is not "part of the job setup" in no way changes the fact that it is necessary to do.

Maybe the job setup isnt correct, but it since it will never be corrected, the allowance needs to be based upon the conditions that exist in the real world rather than some utopian pipe dream where every load is perfect, every car is new and big enough, and every condition is ideal.

I agree that if the route is "blown out", the driver will be forced to sort or do something else outside the methods in order to make the best of the situation.

As I said before, a debate among I.E.'s inside and outside of UPS is how to account for this time. The debate has been going on forever.

If you give planned time for the situation, the driver is then scratch (at least in this element). However, now the bad situation is no longer made visible. No one sees that the management decision has caused ineffective (although necessary) behavior. In addition, how is this controlled. If the route changes, is the driver now under allowed? How do you take this into account?

The other option which UPS chose (and I was on this side of that debate) to not account for the time. The down side of this is that overallowed is no longer a measurement of just the driver. Its the driver AND the job setup handed over from the preload.

I have posted many times that Overallowed works great as a center wide / district wide management tool. Its not good for holding an individual driver accountable to scratch.

I should also point out that when planning the dispatch, overallowed is taken into account. If a route is .5 or 1 hour overallowed, this is added into the planned day to show the dispatched day.

While I will argue that work measurement is pretty accurate for a center, I will never say that every driver should be scratch. Work measurement is generally pointing inefficiencies. It does not point out who caused the problem.

P-Man
 

Pkgrunner

Till I Collapse
Tech,


In the case of select (which is the single largest overallowed area), it is not the driver's fault nearly all the time. The cause is the preload.

Also, interestingly enough, the overallowed in select got worse after PAS / EDD.


P-Man


Isn't this what most drivers on EDD have noticed since its implementation?


I happen to be an under allowed driver the majority of time. I generally run -.5 to -1 on a daily basis depending on the splits and or add cuts I get or give up. I have a good preloader and a very bad trace for most of my route. I hunt a peck throughout my load all day. I figure(proved actually) its still quicker to ping pong inside my package car, than it is to ping pong into and out of canyons several times a day; following the trace that I am told cannot be looped the way I run it because I drive in and out of unit numbers several times.:whiteflag: Granted, when I follow trace, my average under allowed is pretty much the same(-.75 +/-), but my paid day, planned day and total miles all increase while my SPOHR dramatically drops.....I have brought up with my center manager that the focus lately on over/under, for me at least, is not as good of a measurement of productivity as is my SPOHR. They have never pushed the %trace issue or even brought it up with me yet. So, I have not yet had the opportunity to asked him whether he would rather have me drive 150 miles, 95+% trace for an 11 hour plan done in 10 hours paid; or should I drive 125 miles, 60% trace for a 10 hour plan done in 9.25 hours paid. On paper, the "over/under" (and my paycheck:wink2:)is a little better when I burn fuel and drive in and out of boxed canyons several times a day. But really now, which way is actually more productive and less costly.
 
M

Mike23

Guest
Maybe this will put things into perspective for some people.

PCM on Friday, 'When you're done, please call in as we have a lot of people that will be out late'.

To me, that means over dispatched, even though their numbers likely say otherwise. To me that means the numbers failed. To me that means they need to be improved and NOW not later. It's not fool proof, no, but when you know your over dispatching someone and under dispatching another would it not make sense to shift everyone over 1 resi area so everyone is equal? Or am I just thinking to simply here?
 

dannyboy

From the promised LAND
If you give planned time for the situation, the driver is then scratch (at least in this element). However, now the bad situation is no longer made visible. No one sees that the management decision has caused ineffective (although necessary) behavior.

And this is the situation that I find pathetic about the whole situation.

Management will badger a driver for being over allowed each day. To the point of discipline.

But yet by your very words, you have proven that it is a situation that has management causes, and only management solutions.

So why would management take discipline on drivers for something that is management caused?

As far as center managers etc making decisions on their own, I believe that IE should give recommendations to the center team on what to do for the day. IF the manager wants to try something different to experiment with a situation in his center, then he should be allowed to do so. And if the numbers dont bear out what he was wanting to achieve, then go back to IE's recommendation.

But thousands of bright minds are wasted because of the lemming effect. And while they do a good job, and do it well, there are large holes as you have pointed out that have not been addressed. Holes caused by management, holes that are management's responsibility, but yet we are the ones punished for those holes.

d
 

TheKid

Well-Known Member
Okay....

How about the Institute for Industrial Engineering
How about the Institute for Operations Research and Management Science
How about the countless companies that have benchmarked UPS' engineering group to see how to improve theirs.
How about Thomas Friedman in his Chapter on UPS in the book "The world is Flat"

In addition, Under Industrial Engineering is Plant Engineering and Automotive Engineering. P.E. built Worldport which has been deemed a marvel.

Am I brainwashed or are all these sources? Or maybe you are biased.

P-Man
...any of those guys ever deliver a package ?
 

pretzel_man

Well-Known Member
Hello........still waiting for an answer Pretz.

Do you really think your question was relevant?

In case you did not read the previous posts...

I stated that UPS Engineering is seen as world class. That they are recognized as the best of the best.

Someone asked by who? I think posted professional organizations and others that recognize UPS engineering.

Now, do you really think any of those organizations deliver packages? Do you think its relevant?

P-Man
 

TheKid

Well-Known Member
Do you really think your question was relevant?

In case you did not read the previous posts...

I stated that UPS Engineering is seen as world class. That they are recognized as the best of the best.

Someone asked by who? I think posted professional organizations and others that recognize UPS engineering.

Now, do you really think any of those organizations deliver packages? Do you think its relevant?

P-Man
Yes....I do think it is relevant. The time standards they are using now are a joke....they are unatainable. Just a little bio info before I go on. 23 yrs. with UPS....never a warning letter....scratch driver...near perfect attendence....anyway...I always thought the standards were a "guide", but now we have spineless sups. that will not stand up to IE and tell them that they are wrong....instead the sh.t is rolling downhill and landing on some of my friends. That....is relevant to me.
 

TheKid

Well-Known Member
Do you really think your question was relevant?

In case you did not read the previous posts...

I stated that UPS Engineering is seen as world class. That they are recognized as the best of the best.

Someone asked by who? I think posted professional organizations and others that recognize UPS engineering.

Now, do you really think any of those organizations deliver packages? Do you think its relevant?

P-Man
......so how do they know you are any good? If they know nothing about delivering pkgs......how are they in any position to rate you....or is that too much logic for an IE guy ?
 

TechGrrl

Space Cadet
Tech,

I've been out of I.E. for about 10 years. I do know they stopped doing studies for a while, but then started it up again. I don't know how much is observed vs. using data. I do know that they are NOT doing automated time studies (obserless). I think its a combination of the two.

There is one thing that needs to be added to the paragraph above that I bolded.

You have to add to that that the route has a proper job setup. The time study values assume not only proper methods, but that the route is properly loaded.

From what I've seen, this is the single most important area to address. The average driver is over 20 minutes overallowed in select.

Does that mean that the measurement is wrong or that the job setup is not right. As you know, this has been a long debated topic in I.E. If we adjust the allowance, the issue never appears as something to work on.


In the case of select (which is the single largest overallowed area), it is not the driver's fault nearly all the time. The cause is the preload.

Also, interestingly enough, the overallowed in select got worse after PAS / EDD.

Because of these things, I do not think it follows that "If a particular route is consistently too high on off performance, or consistently too low, there is something wrong with the measurement. (again, assuming the driver is skilled and familiar with the area)"

If you add proper job setup to the mix, I would then agree. (I think I'm going to look for my blue card, just for nostalgia.

P-Man

I assumed correct job setup in my comments, but your observation simply adds to my point: an actual that is CONSISTENTLY over or under points out a problem that must be addressed. If setup is a problem, demonstrated by consistently showing as off performance, where then is the ACTION to be taken by the IE function, and the operations managers, to FIX THE DAMN PROBLEM?????

Anyone using statistical process control (which is really what we are trying to do with our time studies) knows that you need to FIX THE PROCESS if you don't like what the process is giving you. So why aren't the IE's and the preload sups and whoever the hell else is in charge of the preload fixing the loading problem????
 
Top