Modeling Isn’t Just For Ships Anymore

Oct 08 2012
OIG Blog Category: Delivery & Collection 10 Comments

The digital revolution has changed communications, and with it, the operations and finances of the U.S. Postal Service. It also has brought deep changes in the way we design networks and analyze systems. Many organizations rely on mathematical modeling to test ideas before they become operational, conserving money and time. The Postal Service, facing limited capital and resources, has also adopted this practice. It is discovering how important these tools are for assessing strategies for designing the future mail network.

The Office of Inspector General has explored some of the main components of the postal supply chain - retail, mail processing and transportation, and delivery – using a systems modeling approach. This approach has allowed the OIG to use objective methods to determine how the network could be redesigned to meet current needs and future demands.

This research also helps us to understand some of the challenges in developing information-based decision models for the Postal Service. A primary challenge in any modeling effort is collecting the necessary information. Without this data, the model cannot fully assess the efficiency of the operations it is modeling, and develop an optimal network solution. Postal information systems can be a complex array of the hundreds of highly varied and specialized information systems that are often developed and maintained under separate contracts. Simplifying this landscape also could enable more insightful analyses to better guide decision makers.

As the Postal Service considers how it can best serve the public through its products and services, modeling efforts can help it to evaluate different proposals for change. As we develop better efficiency standards with more rich data sources, we can not only better evaluate the efficiency of operations and system design, but we can better explore how operations may be changed to meet the needs of new environments.
[poll id="234"]

Comments (10)

  • anon

    It is a waste of money and resources to keep delivering mail that is UNSOLICITED such as: Business Ads, for profit Political Campaign cards Sales Flyers and Coupons 'Current Resident' addressed mail It is a waste of time and a burden to USPS employees who sort and deliver this mail that goes directly into landfills, as tons of paper and ink. There is no way that I can prevent this garbage from being delivered to me.

    Dec 13, 2012
  • anon

    There is Rochelle!! It's called Paper Karma....

    Dec 14, 2012
  • anon

    Sadly, models can, and are manipulated, just as photos are touched up with photo shop Peter is robbed to pay Paul depending on what "model" is being targeted. Once the study is over, it is back to the old way of doing things again, excessive overtime, contracts broken daily, threats and intimidation. Modeling does not help. Employees are overworked and underappreciated. The data you are receiving is inaccurate. When will you stop believing it?

    Oct 14, 2012
  • anon

    Given it is two or three times the size of UPS in terms of carriers, you would expect at least comparable research programs. The UPS has billion dollar modeling operations. What does the USPS have? The discussions and the methodologies surrounding the USPS seem to be a weird combination between being stuck operating somewhere between the 70’s and being ground down by a complicated regulatory environment. I would love to see the USPS bring in a logistics guy from UPS to shake things up, to reframe the conversation on what customers need from the Postal Service.

    Oct 12, 2012
  • anon

    I agree with, "davek", UPS is #1 worldwide in international shipping, we need someone from there logistics team to better access our future of delivery. This has nothing to do with using modeling tools, but I have always felt that the USPS could save hundreds of thousands of dollars by taking away one or two of our paid holidays, really, do we need 11 paid holidays a year?

    Oct 13, 2012
  • anon

    The Postal Service needs to develop workload planning tools that move beyond historical mail volumes but that can be updated based on the actual volumes of mail processed upstream the previous several days. They need to include the amount of work that is necessary for the presorted mail. Perhaps estimating these mail volumes will be easier when Intelligent Mail Barcodes are mandated. If the Postal Service can measure the actual amount of work being done at plants, they can better estimate the work hours that are needed for these tasks, better plan labor needs, and better assess efficiency.

    Oct 11, 2012
  • anon

    The Postal Service is limited by the availability of quality information to its partners. One case in point is the label lists for sorting mail. These lists should be just a few, simple Excel workbooks that are accessible to mailers of all types. For example, these should list all of the five digit ZIP Codes, provide the associated five digit scheme information, dropship delivery unit, and the serving SCF and NDC for dropship mail discounts. There should be simple instructions that accompany these lists. Providing latitudes and longitudes for the delivery units would help to provide a unique identifier for the unit.

    Oct 11, 2012
  • anon

    Here's a modeling algorithm example for you're use.. You can apply any number of variables you wish. http://en.wikibooks.org/wiki/Fundamentals_of_Transportation/Modeling Let's remember one thing though. You've already got the data, and the "breadcrumbs", heck.. you've even got the "lemonade". (These are technical data acronyms, for those readers who don't know.) The simple fact is, as my employment attendance record since 2005 would attest, attendance at the workplace, is not the issue in the transportation, or delivery segment of your model. This raw data is not segregated into a meaningful representation of reality. As, the above URL describes, an example of this would be where one unit is transporting mail, while nine units are static. And, as the wiki description identifies, there exist differences in the "worldview", and the "point of view". As Mr.Jamison correctly identifies.

    Oct 11, 2012
  • anon

    Perhaps you folks ought to take a look at the PRC's most recent Advisory Opinion. The problem may not be that the Postal Service doesn't use modeling, rather, it may be that the Postal Service uses modeling tools for affirmation as opposed to exploration. The folks at L'Enfant Plaza don't appear to be interested in anything other than proving their misguided vision correct. Some of the OIG's own audits and white papers attest to that. Apparently no one wants to deal with the obvious.

    Oct 09, 2012
  • anon

    Modelling may be able to take apart the little fiefdoms which abound in the postal service...The only real problem is that you need accurate numbers and statistics...Management has never let the facts get in the way of what it wants to do. Little fiefdoms always have the proper numbers in spite of reality. Good luck.

    Oct 09, 2012

Share this post


Recent Comments

Monthly Archive