Department of Defense, Defense Logistics Agency, DLA Distribution Anniston, Anniston, Alabama and Local 1945, American Federation of Government Employees, AFL-CIO


The Department of Defense (DoD), Defense Logistics Agency (DLA), DLA Distribution Anniston, Anniston, Alabama (Employer) filed a request for assistance with the Federal Service Impasses Panel (Panel) under the Federal Employees Flexible and Compressed Work Schedules Act of 1982 (Act), 5 U.S.C. § 6120, et seq., to resolve an impasse arising from its determination to terminate the 5-4/9 compressed work schedules (CWS) of bargaining unit employees represented by Local 1945, American Federation of Government Employees, AFL-CIO (Union).
Following investigation of the request for assistance, the Panel determined that the dispute should be resolved through mediation-arbitration with the undersigned, Panel Member Edward F. Hartfield. The parties were informed that if a settlement were not reached during mediation, I would issue a binding decision to resolve the dispute. Consistent with the Panel’s procedural determination, on March 6, 2014, I conducted a mediation-arbitration with the parties at the Employer’s facility in Anniston, Alabama, followed by a post-hearing conference call on March 21, 2014.[1]/ Because the mediation portion of the proceeding failed to result in a voluntary settlement, I am required to issue a final decision resolving the parties’ dispute in accordance with 5 U.S.C. § 6131 and 5 C.F.R. §2472.11 of the Panel’s regulations. In reaching this decision, I have carefully considered the entire record, including the parties’ post-hearing briefs.


The Employer provides distribution services for combat weapons systems, small arms weapons and missile systems for all U.S. military services. It also maintains materiel to support weapons and combat systems, including radioactive and hazardous consumables, major end items and secondary repair parts. Its primary mission is to support the maintenance mission of Anniston Army Depot, the Army's only small arms repair facility, with which it is collocated. The Union represents approximately 203 bargaining unit employees at DLA Distribution Anniston and an additional 4,000 employees at the Anniston Army Depot and the U.S. Army Chemical Materials Activity. The employees affected by the dispute include material handlers, forklift operators, packers, and truck drivers, WG-5 to WG-8. The parties’ Master Labor Agreement (MLA) is due to expire on May 19, 2016.
By way of background, on January 3, 2012, the Employer unilaterally implemented its decision to require employees to work one of three starting times (i.e., 7 a.m., 8 a.m. or 8:30 a.m.) rather than continue on their current 6 a.m. starting time, contending that it was permitted to do so under the parties’ MLA and Locally Negotiated Operating Procedures (LOCNOPS). The Union filed an unfair labor practice (ULP) charge against the Employer with the Atlanta Regional Office of the Federal Labor Relations Authority’s (FLRA). 
On January 23, 2013, the parties executed a Settlement Agreement (SA) which, in essence, required the Employer to reinstate the previous 6 a.m. starting time within 45 days if subsequent negotiations failed to result in a voluntary settlement of the issue. During the 45-day bargaining period specified in the SA, the parties failed to reach an agreement with the assistance of the Federal Mediation and Conciliation Service. Consequently, in accordance with the SA, the 6 a.m. starting time was reinstated on April 22, 2013. Given the disruption to its operations created by the sequestration furloughs ordered by DoD in Fiscal Year 2013, the Employer did not file a request for assistance with the Panel concerning this matter until November 18, 2013.
The sole issue before me is whether the February 4, 2014, finding by the Primary Field Level Activity Commander for DLA Distribution Anniston, upon which the Employer bases its determination to terminate the 5-4/9 compressed work schedules of the 34 affected bargaining unit employees, is supported by evidence of adverse agency impact as defined under the Act.[2]/


1.      The Employer’s Position

The current 5-4/9 CWS, with its 6 a.m. starting time, is causing an adverse agency impact by reducing DLA Distribution Anniston’s productivity and increasing the costs of its operations. Its position is supported by data summarized in “Monthly Performance Metrics” spreadsheets covering the period from August 2010 to December 2011 (when the affected employees started work at 6 a.m.)[3]/ and from January 2012 to April 2013 (when the Employer unilaterally changed the starting times of 34 unit employees to require them to begin their tours of duty at either 7 a.m., 8 a.m. or 8:30 a.m.)[4]/; and by data summarized in “High Priority Performance Metrics” spreadsheets covering the period from April 8 to July 5, 2013 (after the 34 employees reverted to their original 6 a.m. starting time)[5]/, July 8 to August 16, 2013 (during the sequestration furloughs), and August 19 to September 30, 2013 (post-sequestration furlough). The data summarized in these spreadsheets, and presented by the DLA Anniston Deputy Commander at the mediation-arbitration proceeding without contradiction by the Union, provide definitive proof that the staggered starting times resulted in the elimination of “mission failure,” a 24.1-percent increase in customer support, and a 25-percent reduction in labor costs.
Prior to changing the starting times unilaterally in early January 2012 the Employer conducted a “comprehensive study” in October 2011 to identify the root cause of the problem that work generated after 1 p.m. carried over to the next day and did not leave the shipping floor until 25 hours later, causing the installation to consistently fail in meeting the metrics established by the DoD and adopted by the DLA. The study examined the workload generated every hour for the preceding 12 months and the workload volume between the requisition releases (otherwise known as “drops”) occurring daily at 5 a.m., 9 a.m. and 1 p.m. It also “looked at extending work hours” until 6 p.m. “to see what volume of work could be captured.” The study showed that work that was generated after the last drop at 1 p.m. “was basically sitting until the next day to be addressed.”   In addition to this lost productivity, there was an increased direct cost on the Agency in terms of unutilized man hours “as the workforce was not synchronized to the work,” for example, “pickers were around at the end of the workday when you really needed the truck drivers and end stage workers to complete the final requisitions” from the 1 p.m. drop. By “fanning out the compressed work hours” between 6 a.m. and 6 p.m., and making “other administrative revisions,” all requested items were shipped the same day. 
The Employer also has provided post-furlough data showing specifically “the lack of productivity or lack of work that is being accomplished” due to the CWS and all employees leaving at 3:30 p.m. which projected the additional productivity that would result if there were a 5th requisition drop between 1 and 4 p.m. While there was some discussion during the mediation-arbitration proceeding concerning how you calculate the percentage of work not getting accomplished each day with this 5th drop, “what is of paramount importance is that under the current [CWS] such a substantial volume of work is left each day” that the Agency cannot meet its metrics and is experiencing mission failure.    When this volume of work is addressed by a 4 p.m. requisition drop, however, “the Agency is able to meet the DoD requirement.” For these reasons, the Arbitrator should find that the Employer has met its burden of proof under the Act that the current 5-4/9 CWS is causing adverse agency impact and order that it be terminated.
2.   The Union’s Position
The Employer has not met its burden under the Act and, therefore, the Arbitrator should order that the employees’ 5-4/9 CWS with 6 a.m. starting times remain in effect. Its claim that employees should be required to work later in the day to prevent the Agency from experiencing mission failure is contradicted by documents it presented at the hearing establishing that “several bargaining unit employees . . . were actually scheduled to report for earlier tours of duty . . . which ended earlier than their regularly scheduled tour of duty.” If having employees work later in the day results in increased productivity the Employer would not change any impacted employee’s tour of duty to start earlier in the day. Moreover, because the Agency’s requisitions come in 24 hours a day, 7 days a week, the Union agrees with the “underlying logic” that having employees cover more hours allows more work to get done. As witnesses for both parties testified, however, “every requisition/request which is entered into the system after the last drop of the workday and the first drop of the next workday is counted against the operation” unless the Employer requests that DLA adjust the drop cycle, as it does when the facility closes early for a holiday. It is unclear to the Union why tolling the time clock so that, generally, requisitions/requests that come in after duty hours do not count against the operation for determining performance metrics is “impossible,” as the Employer maintained at the hearing.
As to the “comprehensive study” the Employer contends it conducted to identify the “root cause” of its mission failure, and provided to the Union in late 2011, the Union “is not aware of any such business case.” Nor did the Employer present evidence, either before or during the hearing, supporting its allegations that fanning out the starting and stopping times of the 5-4/9 CWS resulted in a 24.1-percent increase in customer support and a 25-percent reduction in labor costs. In fact, when the Union questioned the Employer’s assertion that 17.7 percent of the work in a 24-hour period arrives between 1 and 4 p.m. – the crux of the Employer’s argument for terminating the current CWS – its witnesses “were unable to provide a clear explanation of how they arrived at their stated figures,” throwing into question the reliability of all of the Employer’s data.
On the specific issue of whether the Employer’s data showed that it was experiencing mission failure before it unilaterally terminated the CWS in January 2012, meeting its mission after the termination, and failing its mission once again when the CWS was reinstated in April 2013, the “metrics show only a snapshot and do not provide the missing clear causal relationship between termination of the [CWS] and increased productivity.” In this regard, while the Employer provided metrics for these periods indicating that it was failing to meet standards for “processing other types of work,” some of which also require 24-hour processing to be successful, the Union “finds it incredible” that “its only area of concern” is its most critical requisitions.
Thus, the Employer’s argument is “lacking in substance” because it should be able to show that there was an “across-the-board decline in all areas of performance when employees worked the current” CWS and tours of duty “and a complimentary increase in all areas of performance” when they worked the “fanned” schedule.
The Employer’s data also did not consider a number of other factors that can lead to fluctuations in performance and an inability to meet performance standards, such as higher rates of employee absence around holidays and school breaks, “days when the Cost Center may be closed,” and the fact that “a large number of temporary term employees (25) were lost last year.” In addition, the data did not “account for the very uneven and unpredictable workflow” caused by the drop system, or the unavailability of transportation for an item on a given workday which can result in a “lag in processing time” that counts against bargaining unit employees because “they are unable to load the item they picked from the shelves, processed, identified, counted, weighed, tracked and packed for delivery onto a truck.”
Other flaws in the way the performance data are presented, such as statistics on daily volume during the various time periods in question, rather than by month, make it “impossible to draw clear conclusions” as to whether performance improved when the current CWS was terminated in January 2012. 
Furthermore, the Union notes the “striking difference” between performance figures in May 2013, when the Agency was meeting its standards, and June 2013, when it was not. The current CWS was in effect during both months, so it “would be helpful to know” whether higher levels of annual leave in June contributed to mission failure that month. As the Employer “did not address this issue either in their documentation or at the hearing,” this “provides further evidence” that the Employer has not established that the decreases in performance were caused by the CWS.
In addition, the Arbitrator should consider the Employer’s performance data reports between October 2013 and March 2014, which the Union was unable to present at the hearing because of formatting issues. These reports demonstrate that the unit is “meeting DLA standards for productivity almost every day for which data was provided,” and “clearly debunk” any assertion that the Employer’s performance and ability to provide service to the public “suffered most of the days after return to status quo ante in April 8, 2013.”
In conclusion, the Union believes that any decline in meeting performance metrics for the most critical requisitions between April and September 2013 “is likely due to an issue in operating procedures.” If the issue arises again, however, the Union would be willing to collaborate with management on a “joint analysis of work processes” to improve productivity and service to the public. Moreover, since the parties’ CWS agreement permits the use of a 4/10 CWS, even though “the current metrics show there is no need for any change,” the Union believes such a schedule would be a “viable option” if the Employer can demonstrate a “true need” for employees to cover more hours at the end of the workday.           

           Under § 6131(c)(2)(B) of the Act, the Panel is required to take final action in favor of the agency head’s determination to terminatea CWS if the finding on which the determination is based is supported by evidence that the schedule is causing an “adverse agency impact.” As its legislative history makes clear, Panel determinations under the Act are concerned solely with whether an employer has met its statutory burden on the basis of “the totality of the evidence presented.”[6]/

Having carefully considered the totality of the evidence presented in this case, I find that this Employer has successfully demonstrated that the current 5-4/9 CWS is causing an adverse agency impact by reducing the productivity of the Agency. I find that the DLA data are clear and convincing in proving what common sense and basic logic suggest: If most of the work requisition orders come in between 1 p.m. and 5:45 a.m. the next day, and remain untouched and not handled, how can the Employer possibly meet its mission and the associated performance goals? If no work is done for two-thirds of the day, while requests come in 24/7, won’t any increase in the length of the day result in greater productivity? If most of the workers are allowed to report at 6 a.m. without regard to whether or not they are needed more at another time, is it surprising that there is a logjam that results in a shortage of transportation?
I agree with the conclusion of the study which showed that under the CWS schedule where 34 workers all reported to work at 6 a.m. and left by 3:30 p.m., any work which came in each day after 1 p.m. remained untouched and unprocessed until the next day. I also find that the study supports the conclusion of additional cost to the Agency in the form of lost productivity as the work schedule was disconnected to the needs of the workflow. Specifically, having truck drivers and packers report for work at 6 a.m., hours before they are needed, is as unproductive as having “pickers” around at the end of a workday when what is needed most is drivers and “end stage workers” to finish the work from the last drop of the day at 1 p.m.
The Employer’s decision to unilaterally implement a change in schedule without first bargaining with the Union is not at issue here. That action was addressed in the ULP charge that resulted in the SA that restored the previous CWS. What is significant is that during the period of time in which the Employer spread out (“fanned out”) the starting times of the employees and added start times of 7, 8, and 8:30 a.m., the Agency met its performance goals of on time delivery 14 out of the 16 months and, moreover, shipped the high priority requisitions in 24 hours in 15 of the 16 months.
The Union itself acknowledges the pure logic of the situation on page 5 of its post-hearing brief when it states: “The Union agrees with the underlying logic that being able to have employees covering more hours allows [the Agency] to get more work done, since the [high priority requisitions] do come in 24 hours a day, 7 days a week.”  This statement goes to the heart of the issue. 
I recognize that the Union sees its chief goal in this case to create doubt in, and undermine the efficacy of, the Employer’s demonstration of adverse agency impact, but in my view, rather than spend the time that it does pointing to other factors as playing a contributing role to the improvement in productivity, the Union would be better served by explaining the data to their members and getting them to understand the most important reality: Adding a 5th cycle allows a greater percentage of the work requisitions to be processed on the same day, resulting in an inevitable improvement in productivity.
Helping the Employer to meet, if not exceed, its performance data, may well be the best way for the Union to assure the job security of its members. With the increasing emphasis on cost cutting measures, who is to say that at some point in the future, in a manner similar to the base closures and consolidations, the Agency won’t take a critical look at its 26 similar warehouse/depots around the country and decide which to leave open and which to consolidate based upon the productivity measures?
In reviewing the impressive collection of data presented by the Employer at the March 6 mediation-arbitration hearing, I find the following charts to be among the most persuasive and without a serious rebuttal from the Union:
A.      Page 1, Monthly Performance Metrics, August 2010 to December of 2011
B.      Page 2, Jan 2012 to April 2013 under the “fanned 6am to 6pm schedule”
C.      Page 5:  Missed Workload, Daily Under AFGE Duty Hours
The comparisons on pages 1-2 are obvious. Page 1 shows the amount of time that the Agency is not meeting its major performance metrics under the traditional CWS schedule with the more or less universal 6 a.m. start. The predominantly failing metrics are indicated by the yellow and red colors. By comparison, the results illustrated on page 2 following the Agency’s unilateral implementation of a change in the reporting schedule - the fanned out starting times between 6 and 9 a.m. and the fanned out ending times which ended at 6 p.m. - enabled the Agency to meet its performance metrics consistently during the 16-month period. 
Perhaps the most compelling data is summarized on page 5 - “Missed Workload Daily, Under AFGE Duty Hours, April 8, 2013 to January 31, 2014.” Due to the fact that no new requisitions are handled after the 1 p.m. drop and they continue to accumulate until 5:45 a.m. the next morning, the data show that a staggering 56.81 percent of the workload is waiting for the morning shift when employees start at 6 a.m. Faced with that task, how can the 6 a.m. – 3:30 p.m. crew hope to complete enough of their work on a daily basis? 
The fifth column, titled “Missed Workload,” shows the number of requisitions that come in between 1 – 4 p.m. each day and the percentage that they represent of the daily total. The Union raises a question about the statistical methodology used by the DLA to reach the percentages shown. In its enthusiasm to undermine the credibility of the Agency’s argument and the data on the whole, I believe that the Union has missed the whole point: The missed workload in the 1 – 4 p.m. timeframe constitutes a significant percentage of the daily total. Simply by adding the fifth cycle or drop to address that timeframe, the Agency can make sufficient progress against the overall daily totals. Note that the Agency might demand a second shift to handle even more of the post 1 p.m. work orders; instead, they are asking some of the employees to shift their work schedule by an hour or two.  
In conclusion, I find the Union’s argument sadly lacking on two counts. First, one cannot help but come away with a sense that rather than face the results of the data, the affected employees have become too comfortable in their existing schedule and just don’t want to have to change. How else would one explain the almost total disregard for the Agency’s mission: ”to support the warfighter. When the marine in Afghanistan or the soldier in South Korea needs a firing pin for his firearm or a track for his vehicle, it’s [the Agency’s] responsibility to get that item to him/her within 24 hours of the request.” Does the Union believe that the last sentence should contain the clause “unless it interferes with the warehouse employees going home at 3:30 p.m. every day?”
Second, the Union’s solution to this problem is to suggest that the Agency request the Army/DLA change the metrics system so that the requisitions which come in after all of its employees go home at 3:30 p.m. don’t count against the Agency.  This case is not just about whether an agency of the federal government is meeting its performance metrics. This case is about whether employees in an agency whose performance directly impacts the safety and security of our troops around the globe care enough about that agency’s mission to vary their work times so that critically needed items are returned to the battlefields as quickly as possible.
Pursuant to the authority vested in me by the Federal Service Impasses Panel under the Federal Employees Flexible and Compressed Work Schedules Act, 5 U.S.C. § 6131(c), and § 2472.11(b) of its regulations, I hereby order that 5-4/9 CWS of the affected DLA Distribution Anniston employees be terminated.
Edward F. Hartfield
April 3, 2014
St. Clair Shores, Michigan

[1]/   At the end of the day on March 6, the parties agreed that the Union would survey the affected employees to determine whether there were enough volunteers to work later starting times to meet the Employer’s interests. During the March 21 conference call, the Union reported that no employees were willing to change their 6 a.m. starting times.
[2]/   Under 5 U.S.C. § 6131(b), "adverse agency impact" is defined as: 
(1) a reduction of the productivity of the agency;
(2) a diminished level of the services furnished to the public by the agency; or 
(3) an increase in the cost of agency operations (other than a reasonable administrative cost relating to the process of establishing a flexible or compressed work schedule).
The burden of demonstrating that an existing CWS is likely to cause an adverse agency impact falls on the employer under the Act. See 128 CONG. REC. H3999 (daily ed. July 12, 1982) (statement of Rep. Ferraro); and 128 CONG. REC. S7641 (daily ed. June 30, 1982) (statement of Sen. Stevens). Since the Employer perfected its request for assistance under the Act on February 4, in accordance with the Act’s requirement that CWS termination cases be resolved by the Panel within 60 days, a decision is required by April 4, 2014.
[3]/   During this 17-month period, the Agency met its 85-percent on time metric for its most critical requisitions during only 4 months, and its metric of having the requisition shipped within 1 day during only 3 months.
[4]/   During this 16-month period, the Agency met its 85-percent on time metric for its most critical requisitions during 14 months, and its metric of having the requisition shipped within 1 day during 15 months.
[5]/   After reverting to the 6 a.m. starting time under the CWS “immediately [DLA Anniston] experienced mission failure by not meeting DoD and DLA metrics,” meeting its on time metric for its most critical requisitions on only 16 out of 60 days, and its metric of having these requisitions shipped within 1 day in only 21 days out of 60. 
[6]/   See the Senate report, which states: