March 19, 2008
Six Sigma Receives USPS Stamp of Approval
By Franchetti, Matthew
METHOD DELIVERS RESULTS FOR THE UNITED STATES POSTAL SERVICE The U.S. Postal Service (USPS) has a number of tools at its disposal to ensure that letters and packages get from Point A to Point B on time and in one piece. Web tools and other technological advances allow for easier shipping and tracking of materials, but it was the use of Six Sigma tools that allowed USPS to enhance customer service by improving on-time delivery at an air mail processing center (AMC). Specifically, the largest contributors to delivery failures were examined and reduced at the AMC in Columbus, OH.Annually, the Columbus AMC processes and distributes more than 50 million first class letters that are generated in central Ohio. These letters are transported by as many as 175 daily airline flights from Port Columbus International Airport and Rickenbacker Airport, both of which are located within 15 miles of the city of Columbus. When the analysis began, 8.7% of all letters processed at the Columbus AMC were not delivered on time (based on the organizational goal).
The remainder of this article provides an overview of USPS operations and on-time delivery measurement (which is a key indicator for customer service), a discussion of the Six Sigma method applied and analysis of the results of the project.
USPS delivers more than 213 billion letters and packages annually, operates nearly 37,000 post offices, stations and branches, and delivers to more than 146 million delivery points. This vast network requires a highly developed distribution system to move the mail efficiently and effectively by air and ground.
Transportation expenses for 2006 were $6 billion. Of that amount, 45.8%-or $2.8 billion-was devoted to air transportation. USPS operates 77 AMCs across the country, which process and distribute letters and packages via air transportation, but does not own a fleet of planes. Instead, USPS contracts with commercial airlines and private corporations to provide air transportation.
This study took place at the APC in Columbus, which employs 72 workers responsible for processing and distributing the letters generated in central Ohio area that will be transported by air.
Outstanding customer service is the top priority of USPS. On- time delivery of first class mail from collection box to mailbox is independendy measured by IBM Consulting Services. This includes first class mail with overnight, twoday and three-day delivery commitments based on distance. Most mail that is transported via airplane has a three-day service commitment. This is the primary type of mail processed at the Columbus AMC and is the focus of this study.
Airmail Processing Overview
The Columbus AMC processes more than 50 million letters annually using a combination of automated and manual processes. When a handling unit (which includes letter trays, sacks and large parcels) arrives at the AMC, it is placed onto a conveyor belt that feeds it into a scanning machine. This scanning machine reads a barcode that represents the fivedigit ZIP code to which the handling unit is to be sent for delivery.
A database called the Surface Air Management System (SAMS) is used at each AMC to assign handling units to appropriate air transportation and execute contingency rules using planned and active airline route information.
Based on the handling unit's destination as determined by the ZIP code, it is assigned to a flight in SAMS. Once an available flight is found, the handling unit is assigned to the specific flight, and a dispatch and routing tag is generated and attached. The dispatch and routing tag essentially serves as an airline ticket for the handling unit.
These finalized handling units are then staged in a loading area and collected by the corresponding ground handlers for each airline. The ground handlers then complete a final barcode scan on each item as it is loaded on die assigned flight. These handling units are flown to their destinations, off-loaded by ground handlers at the arrival airport and transported to the receiving AMC to be processed for delivery.
The Six Sigma Team
The Six Sigma team was comprised of the local USPS management team, front line supervisors, mail handlers, an industrial engineer, airline management and two airline ground handlers. This Six Sigma project was completed using the DMAIC method-define the problem, measure performance, analyze the data, improve the process and control the process. The remaining sections discuss an overview of each step.
Table 1. Columbus Airmail Processing Contor at a Glance
Table 2. Combined Test Results
Define the Problem and Measure Performance
The objective of this project was to improve ontime delivery of mail generated in central Ohio to a rate at or above the organizational goal. The process was baselined at the current on- time delivery rate for 50 million letters processed and delivered annually, as determined by IBM Consulting Services. The current process represented a 2.4 sigma level and 187,000 defects per million opportunities.
Analyze the Data
Various Six Sigma tools were applied to identify the vital X's, or root causes, of delivery failures. Numerous reasons contributed to the failures, including incorrect addresses and flight information, poor quality labels, inadequate capacity and cancelled flights.
After developing a process map, a Pareto analysis indicated that airline ground handler delays (26%) and SAMS data system errors (23%) were the top reasons for failures. An analysis of variance using a Chisquare test indicated the following:
* Ground handler delays: By shadowing the commercial airline ground handling crews that load mail onto departing flights, the team found that several delay points existed in the transportation process from the AMC to the airplane. The goal is to transport the mail from the facility to the airport within 30 minutes of the airplane's arrival at the airline staging area. The airlines that serve the Port Columbus International Airport achieved the 30- minute goal for 69% of departing flights.
* SAMS database issues: The team identified incorrect information and lack of real-time monitoring of SAMS, which displays active routes and routes that have closed in the last 12 hours. Several situations were identified where handling units were being assigned to closed flights (flights that had met capacity), cancelled flights or flights that would not provide adequate time to transport the handling units to the plane for loading before takeoff. Delayed flights also were identified as a vital X, as many of these flights delivered the handling units to the arrival processing center dead on arrival to meet the delivery goal. From a three-day survey of SAMS, 11% (50 out of 455) of the flights had incorrect information in terms of capacity, availability or time of departure.
Improve the Process
After identifying the potential vital X's, the hypothesis was tested and validated. Meetings were held with the ground handler crews for each of the major airlines, and a tracking system was developed to record transfer times to the AMC after flight touchdown. Results were shared daily with the airlines' managers.
After meeting with all stakeholders, a standardized process was created and documented. The goal of the standardized process was to "mistake proof any preventable delivery failures. Four weeks of tracking delays revealed that 98% of arriving flights were loaded, and incoming mail was transported to the processing facility within 30 minutes.
Training courses and standard operating procedures were developed for updating SAMS before peak demand periods. Before this time, updates were automatically downloaded from a centralized office located out of state. As a result of the change, a three-day survey of SAMS showed it had correct information for 99% of flights.
As a result of these two process improvements, delivery failures were reduced by 14.3%, from 187,000 to 44,000 defects per million. This reflected an annual savings of $15,000 thanks to a reduction in double handling.
Control the Process
To ensure the process performed within the acceptable limits and fell below the on-time delivery failure threshold, the performance was monitored on an ongoing basis. To achieve "control" status, a p- chart (a tool that tracks defect rates over time) was used.
Data indicated that the on-time delivery rate was above the organizational goal during the two months that followed the test. In addition, ground crew delays were recorded, as well as SAMS updates, which gave the ability to monitor the defect rate continuously.
Monitoring handling crew delays and properly maintaining the SAMS database will enable over 715,000 additional letters per year to arrive within the three-day delivery timeframe. This will increase customer satisfaction, as well as reduce the operating costs associated with double handlings.
Currently, the USPS is expanding its Six Sigma initiatives. Single, well supported projects have been successful, and the company is ramping up to implement an organization-wide movement. To enhance the Six Sigma culture at the USPS, the organization has implemented a training program for industrial engineers. This program recruits recent industrial engineer graduates that have up- to-date skill sets in process analysis and statistics. These recruits are then crosstrained in postal engineering and processing operations and are placed in the field as industrial engineers. Program curriculum involves the addition of a Six Sigma Green Belt project and a certification process.
A PARETO ANALYSIS INDICATED THAT AIRUNE GROUND HANDLER DELAYS AND SAMS DATA SYSTEM ERRORS WERE THE TOP REASONS FOR FAILURES.
WHAT DO YOU THINK OF THIS ARTICLE? Please share your comments and thoughts with the editor by e-mailing [email protected]
By Matthew Franchetti, University of Toledo
Matthew Ranchetti ([email protected]) is an assistant professor in the college of engineering at the University of Toledo in Ohio, where he earned his doctorate in industrial engineering. Ranchetti is a member of ASQ and holds ASQ certification as a Six Sigma Black Belt.
Copyright American Society for Quality Feb 2008
(c) 2008 ASQ Six Sigma Forum Magazine. Provided by ProQuest Information and Learning. All rights Reserved.