Denver Internation Baggage System

tom_stalcup

Lifetime Supporting Member
Join Date
Jan 2004
Location
fresno
Posts
461
I remembered a couple of people mentioning this system in the "best plc ever" thread a while back, so I though I'd post an update to the saga.... Evidently they are losing so much money with the automated system, that by switching back to a manual system they stand to gain $1 million per month.... Insane... Anyone know the details of that system?
yahoo news story
 
No, Tom. I don't know any of the latest details.

However, I do remember very clearly when it came to be and the disaster that ensued. I knew instantly that it was a case of a Hacker trying to be a Process Developer/Process Programmer.

This is not to be confused with your every-day, run of the mill, Process Developer. In general, the typical Process Developer brings things together... once all of the "stuff" is assembled and "co-ordinated", a typical, big-time Process Developer does not get too deeply involved in the details necessary to make a particular process work.

This is NOT to slam those that do in-fact, know what it takes to successfully develop a process from soup-to-nuts.

It is, however, to slam those so-called developers (what we called Process Development Engineers in the computer manufacturing world) that would bring together a package and then dump the job of programming onto some poor schmuck of a programmer. "Specialization" to that degree is absolutely stupid! Have you ever heard of the game called "Telephone"?

But then again, maybe the origninal Process Developer actually did do the programming... if so, considering the results thus far, then... maybe that guy should be "Drummed out of the (PLC) Corp".

I just happened to be flying through Denver when I first heard about the project (I also had a friend in Denver keeping me abrest of the situation). I thought to myself... damn, I wish I was involved in that project... it's nothing more than data management... a piece of cake! It was bound to be a relatively HARD piece of cake, simply because of the shear volume, but none the less, a relative piece of cake... at least with respect to my experience.

And I still believe that to be so!

The key to any large project is MODULARITY! Whether it is sensor activity occurring or data manipulation at a particular activity, it boils down to knowing what you have and what you plan to do with it. This occurs at the modular level.

Damn... I really, really wish I was the developer/programmer for that project.

Yeah... ego... but then, some of us have what it takes.

DAMN! I REALLY, REALLY, REALLY WISH I WAS INVOLVED IN THAT PROJECT!

Denver would have been pleased.

It might have been slower than expected at the beginning, but it would have been as effective as required. Eventually, considering how fast airport traffic is, it would have become as efficient as the hardware would allow within a week or so.

And, oh, yeah, at this point, a little MGD doesn't hurt in the self-appraisal.

Scratch that... I have no doubt that I could have done a better job.
 
My 2 cents:

From what all I've read on the subject, the real issue seems to be with the mechanics of handeling (via total automation) an infinate variable of different sized, weight, shape etc..bags. If all the boxes where about the same configuration, sorting them wouldnt be a problem. Thats why FedX, UPS, etc.. have standards for packaging.
 
Logplan must have been in on the remedial measures; I don't think any of BAE's original system still exists.

DIA has so many failure factors that it's a major subject of studies in complexity theory, let alone automation theory.

Even the stopgap measure, which used more conventional PLCs and Wonderware, was done without organization; the incoming baggage handler contractor used A-B controllers, the outgoing baggage handler contractor used Square-D controllers. Brilliant !

I don't know exactly who was the vendor on the PLC side of things at DIA in the beginning; I just know, and am very glad, that it wasn't ME.
 
Stuff like this happens all the time but in smaller scale. The clueless hand the job to "super-duper" .Net expert (heck, PLC must be easier than VB!). And to make things work.. patch after patch of little VB applet has to be used with PLC as dummie IO. DIA is not the first nor will it be the last automation fiasco.
 
I'm sitting here thinking about this and wondering how complicated cant it be?

If each bag has an RFID tag, then toss it onto a conveyer. at each carosel it's scanned, and if it's for that carosel, push it.

I see two conveyers; one for outgoing (to the gate/plane) and one for incoming, (to the baggage pickup). It's a simple matter of pushing it off the conveyr at the right place, which cant be that diffucult.

The only real problem I can think of would be dealing with the mechanics, like straps, and soft luggage, (like duffelbags) getting caught up in the works.
 
ah-hem.....

Attention... Denver International Airport...

If you want a working system, I mean, a really working system, then you and I should talk. My finesse is in details... that lack of detail control is what's killing you!

Sometimes it takes a little time for me to get it nailed... the easy stuff is "right now"... the impossible takes a little longer. Your situation is somewhere between.

Can you afford the time? After all, how long has it been under the current, and the previous, so-called solution?

Denver, I would LOVE to help you get this right... even if it is one of those damnable older AB's.

Hello? Dever Tower? Are you there?
 
Good point...

harryting said:
Stuff like this happens all the time but in smaller scale. The clueless hand the job to "super-duper" .Net expert (heck, PLC must be easier than VB!). And to make things work.. patch after patch of little VB applet has to be used with PLC as dummie IO. DIA is not the first nor will it be the last automation fiasco.
And I think this is the same thing Terry was getting at...
Terry Woods said:
The key to any large project is MODULARITY! Whether it is sensor activity occurring or data manipulation at a particular activity, it boils down to knowing what you have and what you plan to do with it. This occurs at the modular level.
The best project are well planned... absurd to use a PLC only as dumb I/O, PLCs are so much more powerful.
 
Daniel,

I dont think a PLC is the solution or problem. Think about tracking thousands + bags at one time. This will reqire a pretty good sized database, which a PLC is not designed for. Think about it; a huge central database will have to report to each line what bags to expect, and each line will have to report back what bags are where. This is definately a network job with a central PC control.
 
Missed the point...

I probably did not state that clearly... But as both I had quoted stated, its planning that is important and as Terry stated its using the right equiptment for the right proccess... In short the PLC should not be used as dumb I/O, it could infact control most if not all of the package movements, alarms, jam problems, etc... The security and Database software/hardware obviously needs to be well selected to give the PLC the proper information...
 
Try as I might I cant find any articals that explain WHAT is wrong with the system. Is is ripped up baggage & jambs, or is it dumping the wrong bags in the wrong places? Are you going to Flordia, and your bag's going to Tibet? or is it that your bag looks like a bag of shreaded wheat when you do get it?
 
Ken Roach said:
You guys are underestimating by an order of magnitude.

Ken, I'm not so sure of that. Maybe the problem is that the project was over complicated to begin with? It seems that an attempt might have been made to re-invent the wheel, and out came a cube.
 

Similar Topics

Does anyone deal with a Gardner Denver 3 stage compressor, which is controlled by a Toshiba PLC. A company had the Toshiba PLC burn up, and they...
Replies
0
Views
1,688
I wonder if there will be any posts concerning the Heathrow Terminal 5 baggage system after the reported 400,000 hours of software design ? Anyone...
Replies
32
Views
10,682
Back
Top Bottom