Sunday, May 2, 2010

Optimizing the Postal Service Network and 6-day Delivery

One of the benefits of the 6-day to 5-day proceeding is that it provides an enormous amount of data relevant for trying to understand the Postal Service's strategy to control costs.  A recent response by the Postal Service to an interrogatory by Douglas Carlson provides the first nationwide picture of the Postal Service's effort to consolidate sortation on Saturday among fewer facilities.

The Postal Service can consolidate sortation of collection mail that is collected on Saturday because it has at least 12 hours of time to transport the mail for sortation at a distant facility and get that mail back for final sortation at the destination facility.  The list of facilities that Douglas Carlson elicited from the Postal Service identifies 139 facilities that do not sort on Saturday.  A cursory look at the distance between facilities that sort mail only Monday through Friday and facilities that sort mail within that facility's territory on Saturday suggests that the Postal Service could be significantly more aggressive in consolidating sortation on its slowest day of the week.

The question as to how many facilities are needed to sort Saturday collection mail is determined by the time it takes to sort this mail, the available capacity in the canceling and originating sortation operations and the time it takes to transport the mail from one facility to another.  The Postal Service has shown that it can handle its service commitments by consolidating Saturday sortation at facilities more than 100 miles away from the facility that sorts the mail the other 5 days.  Examples include:
  • Rapid City SD to Sioux Falls, SD - 348 miles
  • Truth or Consequence, NM to Albuquerque, NM - 149 miles
  • Green Bay, WI to Milwaukee, WI - 116 miles

If it is physically possible to consolidate sortation into a network that transports mail collected on Saturday 100 miles or more in South Dakota, New Mexico, and Wisconsin, then it should be possible to do the same in all other regions of the United States.  

Now this raises a second question is, "does it reduce cost of operations?"  Is the reduction of processing costs greater than the increase in transportation costs?  Given that the Postal Service can cost-justify consolidating operations from Rapid City to Sioux Falls and between Green Bay to Milwaukee, that I would expect that similar analysis in Pennsylvania might support consolidating all Saturday sortation in no more than three facilities in Pittsburgh, Harrisburg, and Philadelphia and it may even be possible to consolidate Saturday collection mail sortation in Pennsylvania in two facilities. Similar examples can be constructed in other states. 

What this means for the 6-day to 5-day proposal is that the Postal Service could most likely provide 6-day service using fewer facilities and at a lower cost than it does today.   Choosing to become more efficient has a major drawback for postal management; it creates more excess career employees.    Excessing these employees would have up-front costs in terms of retirement incentives and severance pay.   The Postal Service does not have the cash to cover these costs.  In addition to increasing immediate cash needs, excessing career employees create additional political headaches that the Postal Service does its best to avoid.

The possibility that facility consolidation could reduce the need to cut one day of delivery also creates a conundrum for postal labor.  Postal unions are all opponents of the 6-day to 5-day proposal, however, it is unlikely that their filings before the Postal Regulatory Commission will identify how the Postal Service could change its operations to reduce costs and union jobs if that was needed to keep 6-day delivery service.  In many ways, unions may find that 5-day delivery is preferable to consolidating the processing network as it minimizes the number of career jobs that are lost.

Of all stakeholders with an interest in the 6-day to 5-day proposal, clearly the Postal Service's customers have the most to lose by the less than aggressive strategy consolidating the operating network.   If 6-days of delivery are preferred by the postal market, it appears that the Postal Service did not act aggressively enough to reduce capacity and costs to keep 6-day delivery financially viable.   If the market is indifferent to 6-day or 5-day delivery, or if financial losses are so great that 6-day is unlikely to ever be financially viable, then customers still lose as the less than aggressive effort to consolidate facilities has resulted in higher costs that the Postal Service needs to recover through an exigent rate case it will file this summer.  

Congressional hearings are not a particularly good forum for analyzing the consolidation of the postal processing network and the impact of any consolidation strategy on postal costs and the rates that its customers pay.  Yet testimony from Michael Coughlin, and the Government Accountability Office indicted that such an analysis needs to be conducted by an independent entity.   It may be worthwhile for Congress to ask a government entity other than those that have previously provided testimony to the relevant committees to conduct such an analysis.  For example, the Department of Transportation and in particular the Volpe National Transportation System Center has the capability of managing such a study.  Such a study could help Congress evaluate all of the legislative changes that the Postal Service requests as well as whether GAO's proposal of a BRAC type commission is needed.


Anonymous said...

An article obviously written by someone who know little about COSTS. This inculdes transportation cost, required manpower to do who he suggest or the fact that the postal services greatest cost is labor. There are only 2 ways to reduce labor cost.
Become more efficient(productive) or reduce manpower.
The writer of this article seems not to know about costs..

Anonymous said...

I completely agree with you. However, as I have had a hand in some of these proposals, the service standard differences for each 3 digit ZIP area are the key constraint. We have yet to implement an AMP with a service degradation. HQ has refused any that show significant savings yet require a degradation of the service standard.

Alan Robinson said...

I can understand the concern about changing service standards, but given the time available to transport mail when there is the extra day between pick-up and delivery for Saturday collections, there should be few consolidations that could be considered that would have service impacts.

Also, it may be worth thinking about a network realignment as a balancing act. The balancing act should be made based on the proportion of volume affected not the proportion of origin-destination pairs. It is possible that some origin-destination flows could see a lower service standard, but have so little volume that the impact on postal customers is minimal. While consolidations, could speed the delivery of mail from printers to households, a step that would improve the value of the mail product and keep it relevant.

johnD said...

For full disclosure what parcel companies, postal services do you work for at present. I am just wondering if you views are slanted for or against the Postal service.

Alan Robinson
has over twenty-five years experience working on the regulatory, policy, marketing, human capital and management issues associated with changes in competition within transportation, parcel delivery and postal markets. His clients have included trade associations, transportation companies, postal operators, parcel carriers, shippers, and suppliers to transportation companies and postal operators. He is the author of the blog Courier Express and Postal Observer.

Alan Robinson said...

The opinions on this blog are my own. I do not work for any stakeholder with an interest in postal issues. I have managed the development of network optimization models similar to ones that would be used to determine the number and location of facilities that the Postal Service requires.

uncommonsense said...

First, I believe you are making some very apples to oranges comparisons here.
The efficiencies gained in consolidating a low volume day in a low population area are significantly greater than those gained by doing the same consolidation in a higher population area. If there is enough volume to support a couple machines and a crew for a tour of canceling and outgoing processing there isn’t nearly as much efficiency to be gained through consolidation as in rural areas where there isn’t even enough volume to support a couple machines and a crew.
The fact that it is economically viable to transport one truck of mail 200 miles in MT or NM on Saturday for consolidated processing has little relationship to whether or not it is economical to transport several truckloads of mail from the Lehigh Valley 100 miles to Philly for processing on Saturday. There is already probably enough volume in the Lehigh Valley to run an efficient outgoing operation. Trucking that mail to Philly would likely yield only marginal if any gains in processing efficiency and would definitely lead to higher transportation costs.
Second, mail processing has become so much more efficient over the years that the efficiencies to be gained through consolidation probably are not big enough to make a significant dent in postal operations. Looking at the most recent 10Q the USPS reports less than 61 million hours used in mail processing or about 20% of overall hours. Delivery accounted for about 148 million hours or nearly 50% of overall hours. Delivery is becoming an ever larger percentage of overall USPS costs and as such any efficiency gains made in delivery have a much bigger impact on postal operations than efficiency gains in mail processing.
If 5 day delivery cuts delivery costs by 10% it would decrease postal costs by 5% directly plus it would have some impact on reducing mail processing costs. If network consolidation cuts mail processing costs by 10% it would decrease postal costs by 2% directly and likely indirectly increase delivery costs as the farther the delivery unit is from the processing unit the more chances there are for problems to occur in the final delivery products arrival time and condition. Because delivery is so much larger than processing this could eat up most of the 2% in savings.
Looking at the 1st quarters numbers, compared with 5 years ago volume has decreased 19%. Mail processing hours have decreased 33% and delivery hours have decreased 10%. Mail processing productivity is up significantly over the last 5 years despite the alleged inefficiencies created as the network becomes too large for the volume of mail. Delivery productivity has decreased and is eating up an ever larger percentage of the USPS budget. Delivery has become 2 and a half times the size of processing. That is why the impact of changes in the processing network cannot be directly compared to changes in the delivery network.