Inventory Management: Handheld Missiles Are Vulnerable to Theft and Undetected Losses (Letter Report, 09/16/94, GAO/NSIAD-94-100). Among other discrepancies, 40 deadly Stinger missiles shipped to the Middle East during the Persian Gulf War could not be accounted for when GAO visited military storage sites to check the inventory of handheld missiles--the Stinger, Redeye, and Dragon. In all, inventory records differed from GAO's physical count by thousands of missiles. The military services do not know how many of these missiles they have in their possession because they have not established systems to track the missiles produced, fired, destroyed, sold, and transferred by serial number. Lax military oversight and recordkeeping have left these missiles, which are in demand by terrorists and drug dealers, vulnerable to theft. --------------------------- Indexing Terms ----------------------------- REPORTNUM: NSIAD-94-100 TITLE: Inventory Management: Handheld Missiles Are Vulnerable to Theft and Undetected Losses DATE: 09/16/94 SUBJECT: Inventory control systems Missiles Military inventories Records management Internal controls Reporting requirements Facility security Larceny Federal property management Property losses IDENTIFIER: Persian Gulf War Stinger Missile Redeye Missile Dragon Missile ************************************************************************** * This file contains an ASCII representation of the text of a GAO * * report. Delineations within the text indicating chapter titles, * * headings, and bullets are preserved. Major divisions and subdivisions * * of the text, such as Chapters, Sections, and Appendixes, are * * identified by double and single lines. The numbers on the right end * * of these lines indicate the position of each of the subsections in the * * document outline. These numbers do NOT correspond with the page * * numbers of the printed product. * * * * No attempt has been made to display graphic images, although figure * * captions are reproduced. Tables are included, but may not resemble * * those in the printed version. * * * * A printed copy of this report may be obtained from the GAO Document * * Distribution Facility by calling (202) 512-6000, by faxing your * * request to (301) 258-4066, or by writing to P.O. Box 6015, * * Gaithersburg, MD 20884-6015. We are unable to accept electronic orders * * for printed documents at this time. * ************************************************************************** Cover ================================================================ COVER Report to the Chairman, Committee on Governmental Affairs, U.S. Senate September 1994 INVENTORY MANAGEMENT - HANDHELD MISSILES ARE VULNERABLE TO THEFT AND UNDETECTED LOSSES GAO/NSIAD-94-100 Inventory Management Abbreviations =============================================================== ABBREV Letter =============================================================== LETTER B-256115 September 16, 1994 The Honorable John Glenn Chairman, Committee on Governmental Affairs United States Senate Dear Mr. Chairman: In response to your request, we inventoried the military services' most sensitive (Category I) missiles. The reported loss of control over these missiles during the Persian Gulf War had raised concern that some of these missiles may be missing. We visited all 78 land-based storage sites, counted each missile container, opened a sample of these containers, and recorded pertinent identifying information such as serial numbers. We excluded missiles stored aboard ships and provided under the Foreign Military Sales Program. We compared the number of missiles counted at each location to two different levels of records: (1) the item managers' records and (2) the records at that location or base. BACKGROUND ------------------------------------------------------------ Letter :1 Category I missiles--the Stinger (see fig. 1), Redeye, and Dragon--are handheld, accurate, lethal, and in most cases ready to fire. Figure 1: The Stinger Missile (See figure in printed edition.) Since 1970, several hundred thousand of these missiles have been produced and issued to the military services, and thousands were sold to other nations through the Foreign Military Sales Program. Because the Stinger and the Redeye can destroy aircraft in flight and the Dragon can pierce armor, they are in demand by terrorists, insurgents, and drug dealers. The Army and the Marine Corps are the primary purchasers of Category I missiles; consequently, we focused our review on their inventories. In 1991, the Defense Department's Inspector General inventoried 60 percent of the Army's and the Marine Corps' Stinger missiles. The Inspector General could not account for 188 missiles and concluded that the two services' inventory records were inaccurate and reporting procedures were ineffective. The Inspector General considered the inventory variances to likely be the result of paperwork problems. The last page of the report contains a list of the Defense Department, the Army Audit Agency, and our reports published over the years on the control of and accountability for Category I missiles, ammunition, and explosives. RESULTS IN BRIEF ------------------------------------------------------------ Letter :2 Many serious discrepancies in the quantities, locations, and serial numbers of Category I missiles indicate that the services have poor oversight and record-keeping of these lethal weapons. Further, the services do not know how many Category I missiles they should have in their possession because they did not establish systems to track the missiles produced, fired, destroyed, sold, and transferred by serial number. We therefore could not determine the extent to which missiles may be missing from inventory. Service inventory records differed from our physical count by thousands of missiles. In addition, physical security measures are not uniformly applied at all locations where these missiles are stored. Moreover, during the Persian Gulf War, inventory problems complicated accountability of the missiles. According to law enforcement officials, thefts of missiles from U.S. inventories have been alleged numerous times over the years, but no such thefts have been confirmed. The poor oversight and record-keeping of Category I missiles, however, lead us to conclude that these missiles have been and remain vulnerable to theft or other undetected losses. PAST PROBLEMS HAVE BEEN REPORTED ------------------------------------------------------------ Letter :3 Some missile inventory problems can be traced to the Persian Gulf War. Although the Army's item managers\1 had knowledge of Army missiles within the wholesale system and could direct wholesale shipments to the Persian Gulf, they either lacked or believed they had inaccurate information on missiles that were shipped at the unit level. Some Army combat units deployed to the Gulf with more than their authorized levels of ammunition, including Category I missiles. In addition, we were told that several units disposed of their inventory records before deployment to the Gulf, and at one supply point, both hard copy and automated inventory records were allegedly destroyed. The Marine Corps knew how many missiles it sent to the Gulf because it sent all it had. In the Persian Gulf, the combat theater commander relaxed administrative requirements, as permitted by regulation, so that weapons could be quickly provided to support military operations. According to Army officials, at the entry port, combat units took what they wanted. In addition, some missiles were transported unguarded on trucks driven by third country nationals, and some ammunition sites were wide open. According to one Army official, due to the lack of accountability, it would be "pure luck" if no missiles were lost. The Marine Corps inventoried missiles at entry ports to ensure that units received the appropriate number of missiles, and Marine guards accompanied each vehicle carrying missiles. Although some units signed for their missiles, particularly Stingers, others did not. In addition, units frequently shared ammunition. According to one Marine official, missile accountability was lost at this point, if not earlier. To expedite troops' return to the United States at the end of the war, units were allowed to turn in weapons, ammunition, and equipment without documentation. Unauthorized items, such as enemy weapons, could be placed in "amnesty" boxes; some Stinger missiles were also placed in these boxes. In addition, according to Army officials, a load of Stinger missiles was found unguarded on the side of a road. -------------------- \1 Item managers are responsible for Category I missiles while they are at or en route to and from a depot and for reporting missile inventory levels and locations for combat deployment and sustainment requirements. Historically, the item managers have managed missiles by quantity and not by serial number. SERVICES RECONCILE MISSILE INVENTORIES AT WAR'S END ---------------------------------------------------------- Letter :3.1 The Army and the Marine Corps reconciled the missiles sent and returned from the Persian Gulf with reported combat use and concluded that no missiles were missing. According to the Army's analysis, 6,373 Stinger missiles were shipped to the Gulf, and the same number of missiles were returned. The Army said it did not fire any Stingers in combat or training while in the Gulf. While the Army could reconcile the missiles it was told were sent with those returned by quantity, it could not reconcile by serial number. A comparison of serial numbers showed that 40 of the missiles sent to the Gulf were not returned to the depot, other Army locations, or the other services. The Army does not know where these missiles are. In addition, the Army could not determine ownership for 1,400 of the missiles returned because it did not have serial numbers for all the missiles sent to the Gulf. The item manager provided a list of these serial numbers to all Army units and to the other services; however, none of the missiles were claimed. The Army's reconciliation also does not include 106 missiles taken by a unit to the Persian Gulf against instructions and therefore not recorded as having been sent to the war. The unit returned the missiles to the depot after returning from the Gulf War. It also does not account for a Stinger missile that was damaged and then destroyed in theater. According to a Marine Corps official, the Marine Corps shipped 3,754 Stinger missiles and 7,485 Dragon missiles to the Persian Gulf. All were returned except for 7 Stingers and 160 Dragons, which were reportedly fired during the war. However, the Marine Corps cannot be certain that all unfired missiles were returned because it did not compare the serial numbers of missiles shipped with those returned. Furthermore, the Army's reconciliation includes a missile that is also claimed by the Marine Corps. THE SERVICES DO NOT KNOW HOW MANY MISSILES THEY SHOULD HAVE ------------------------------------------------------------ Letter :4 Our review showed it is impossible to accurately determine how many missiles are missing at the item manager or storage level because the services did not establish effective procedures to determine what should be in their inventories. Such a procedure would compare the number of missiles produced less the number fired, destroyed, sold, or transferred with on hand inventories. Although the services have collected this information over the years, they stated it is inaccurate and incomplete and therefore unreliable. The services stated that they cannot correct the situation because they cannot reconstruct the data that would be needed. ITEM MANAGERS' RECORDS ARE INACCURATE ---------------------------------------------------------- Letter :4.1 Oversight responsibility for each type of Category I missile has been assigned to a single individual or organization within each service. However, responsibility for control, security, and accountability of these missiles rests with those organizational entities having physical custody of them. This means that item managers are expected to know, at any point in time, how many missiles are in inventory and where they are located, but are dependent on many others for this information. Item managers told us that reporting this information is a problem because some units fail to report or inaccurately report missile inventories. Oversight organizations do not have the authority to direct compliance with reporting requirements. They are also dependent on those having custody of missiles to conduct required and unscheduled physical inventories and to report any adjustments to them. The services referred us to their item managers for information on how many missiles were in inventory and where they were located. The item managers told us they would have difficulty providing missile quantities and storage locations and would have to contact combat units and storage locations directly to get accurate information. Nevertheless, item manager records differed substantially from our physical inventory count of the missiles (see table 1). Table 1 Comparison of Our Inventory With Item Manager Records, by Missile Type Service Missile\a Our inventory records Difference\ --------------- ------------- ------------- ------------- Stinger 36,216 28,484 7,732 Dragon 40,359 50,103 9,744 Redeye 7,983 2,753 5,230 ------------------------------------------------------------ \a We did not include missiles on ships at sea. The following examples show how these differences could be so large: The item manager's record of missiles at one major depot showed 7,370 in the inventory; we counted 12,426. In Europe, item managers' records indicated that 22,558 Category I missiles were on hand; we counted 20,373, a difference of 2,185. At a manufacturing facility, we counted 953 more missiles than the item manager told us were there. This variance resulted in part because contractor employees deleted 130 missile serial numbers from the database to make it match the number of missiles they believed were in the two magazines. The services have recognized that they must control Category I missiles by serial number. The Navy and the Marine Corps item managers began controlling missiles by serial number in 1990 and late 1992, respectively. Since the Persian Gulf War, the Marine Corps has inventoried its missiles twice--once by quantity and more recently (Nov. 1992) by serial number. Although the Army's Stinger and Dragon item managers are working on obtaining control by serial number, the Redeye item manager is not because the missile is being phased out. The Army has not conducted a worldwide inventory to establish an inventory baseline by serial number. STORAGE SITE RECORDS ARE INACCURATE ---------------------------------------------------------- Letter :4.2 Missile inventory records at the storage sites contain discrepancies in quantities and serial numbers and problems with national stock numbers. Of the 78 missile sites we visited, 31--or 40 percent--did not have accurate records of their on-hand inventories. For example, after we had finished our inventory at one storage site, personnel called to tell us that they had located 70 more missiles in another magazine. Table 2 shows the number of sites that had inaccurate quantities, by type of missile. Table 2 Inaccurate Quantities at Missile Storage Sites Percent of Sites with sites with Sites storing inaccurate inaccurate Missile\a missile quantities quantities --------------- ------------- ------------- ------------- Stinger 44 18 41 Dragon 57 17 30 Redeye 29 6 21 ------------------------------------------------------------ Note: Some of the 78 sites store only one type of missile; others store more than one type. \a We did not include missiles on ships at sea. We found either more or fewer missiles than records indicated were on hand at 31 different sites. The differences ranged from one missile at several sites to hundreds at other sites (see table 3). Table 3 Variances at Sites With Inaccurate Records Site Missile Variance -------------------- ------------------ ------------------ A Dragon 423 B Stinger 77 C Dragon 1 D Dragon 158 E Stinger 10 F Stinger 3 Redeye 5 G Stinger 1 H Stinger 4 Dragon 598 I Stinger 1 J Stinger 2 K Stinger 8 Dragon 1 Redeye 1 L Dragon 72 Redeye 2 M Stinger 10 N Dragon 5 O Stinger 41 P Stinger 142 Q Stinger 142 R Stinger 70 Dragon 22 S Dragon 9 T Redeye 49 U Dragon 1 V Stinger 63 Dragon 42 W Redeye 4 X Stinger 83 Dragon 50 Redeye 4 Y Dragon 90 Z Dragon 3 AA Dragon 2 BB Stinger 10 Dragon 1 CC Stinger 1 DD Dragon 4 EE Stinger 18 Dragon 62 ------------------------------------------------------------ We did not attempt to reconcile the differences with inventory personnel, and the differences may result from record-keeping problems. The possibility of undetected loss, however, remains. Although most of the missile containers we examined showed serial numbers, national stock numbers, and the Defense Department identification codes, as required by regulation, we found duplicate, missing, or illegible serial numbers as well as other problems at many of the locations we visited. Other serial number problems found at many locations were bar codes but no stenciled serial numbers; stenciled serial numbers that were not legible; bar codes that did not match the stenciled serial numbers; two different stenciled serial numbers; serial numbers handwritten in chalk; serial numbers written on paper tags; and no identification because the tags or bar codes had fallen off, or the serial numbers written in chalk had been partially or completely erased. Confusion over which national stock number is assigned to a missile configuration\2 also resulted in some units dropping missiles from their reports. Missiles were also dropped from inventory records when parts were removed for testing, maintenance, or inspection, increasing the possibility of errors and resulting in loss of oversight. -------------------- \2 The national stock number identifies the missiles' configuration. The Stinger has 22 different national stock numbers. MISSILE AND LAUNCHER TUBE MISSING FROM CONTAINER ---------------------------------------------------------- Letter :4.3 At each location we visited, we opened about 1 percent of the missile containers to see if a missile was enclosed. We found that (1) one missile was missing from its container, (2) one missile was missing its launcher tube, and (3) some containers had serial numbers that did not match the enclosed missiles. Depot personnel believed the missing missile had been destroyed, but they had no documentation to support their belief. They also said the missing launcher tube had been sent to a testing facility. They could not explain the mismatches in serial numbers. When we pointed out a mismatch at one location, a maintenance man quickly spray-painted the serial number on the container to match the number on the missile inside. No one verified, however, whether the changed serial number matched another live missile. SECURITY CONTROLS ARE NOT UNIFORMLY OBSERVED ------------------------------------------------------------ Letter :5 Although most of the missile storage sites we visited provided adequate barriers to outside intrusion, some had not implemented required security measures. For example, several sites used only one key to open a missile magazine, although two keys controlled by separate individuals are required for each magazine. Other sites had no fence surrounding the magazine area, even though fencing topped with barbed or razor wire is required. At still others, the alarm systems did not work. In addition, some missiles packaged for rapid deployment were stored on flatbed trucks or trailers parked in secured areas, while others were stored in magazines. Missiles undergoing environmental testing were also stored outside. Category I missiles are required to be in a magazine controlled by a guard or locked and monitored by a video camera, and the magazines must have alarms that can be deactivated before entering. According to service officials, some of these exceptions were covered by waivers. Security regulations and procedures directed at employee theft were not uniformly applied at all locations. Security guards routinely and thoroughly inspected unfamiliar vehicles entering or leaving the ammunition area. They checked for appropriate identification, opened trunks, and looked in the back seat and under the vehicle. Security guards, however, do not routinely inspect all vehicles entering or leaving ammunition storage areas. Guards are only required to conduct spot checks based on guidance provided by the local commander. At one missile storage location, we witnessed a spot check where a guard found and removed two new boxes of small arms ammunition that was hidden in a trash truck. The guard permitted the driver to pass through the gate without questioning and did not file an incident report but did make an appropriate notation in his log. Not all trash trucks or other vehicles that could easily conceal missiles are inspected when leaving an ammunition storage area. We previously reported that military inventories remain more vulnerable to employee theft than to outsider intrusion.\3 -------------------- \3 Inventory Management: Strengthened Controls Needed to Detect and Deter Small Arms Parts Thefts (GAO/NSIAD-91- 166, July 17, 1992) and Small Arms Parts: Poor Controls Invite Widespread Theft (GAO/NSIAD-94-21, Nov. 18, 1993). ALLEGED THEFTS OF CATEGORY I MISSILES ------------------------------------------------------------ Letter :6 According to law enforcement officials, thefts of missiles from the Defense Department inventories have been alleged numerous times over the years. Allegations are examined for merit and investigated when warranted. By agreement, the Defense Department is to notify the Bureau of Alcohol, Tobacco, and Firearms within 72 hours of the confirmed loss of a missile. According to Bureau officials, the Department has never reported such a loss. In addition, other law enforcement agency officials said that they had no data on confirmed thefts. We tried to follow an allegation (Apr. 1992) made regarding the theft of Stinger missiles from an Army storage depot. According to the Bureau, an informant said that nine Stinger missiles were going to be diverted/stolen and had been moved to a location that would facilitate their easy removal from the depot. The informant also indicated that as many as 20 other Stingers had been diverted/stolen in this manner. The Bureau immediately provided this information to the Defense Department, and an Army investigator found the missiles in the location identified by the informant. When we asked about the results of the investigation, Department officials said that they vaguely remembered the incident and that inventory personnel had plausibly explained why the missiles were in that location and had confirmed that no missiles were missing. The Defense Department, however, could not provide any investigative documents to confirm that an investigation actually took place. Inventory personnel told us that the Defense Department does not normally conduct independent inventories of its missiles every time an allegation is made because it would be too costly and time-consuming. Further, missile containers are not opened during inventory to ensure that they contain missiles. RECOMMENDATIONS ------------------------------------------------------------ Letter :7 We recommend that the Secretary of Defense direct the Secretaries of the Army, the Navy, and the Air Force to authorize Category I missile oversight organizations to enforce missile reporting requirements and to conduct unscheduled independent inventories at depot, post, base, or unit level missile storage sites; conduct independent worldwide inventories of Category I missiles by serial number to establish an accurate baseline of existing missiles; establish procedures to track, document, and report additions to and deletions from these new inventory baselines; establish procedures to include a random sampling of missile containers during inventories to ensure that they contain missiles; reemphasize employee security procedures so that they are consistently and uniformly applied to all individuals entering and leaving missile storage areas; and reexamine the current security policy that permits less than full inspection of vehicles, such as trash trucks, that could easily conceal missiles when leaving ammunition storage areas. AGENCY COMMENTS AND OUR EVALUATION ------------------------------------------------------------ Letter :8 The Defense Department generally agreed with our findings and accepted our recommendations. The Department recognizes that improvements are needed in serial number controls of Category I missiles and agreed to strengthen inventory accountability by (1) reviewing guidance to make inventory management among oversight organizations as uniform as practicable; (2) completing a serial number inventory of Stinger, Dragon, and Redeye missiles; (3) maintaining a permanent record to account for missile ownership and use by serial number; (4) opening a sample of missile containers during inventories; and (5) reemphasizing policies for controlling access to missile storage areas and the deterrent value of consistent screening of vehicles entering and leaving storage areas. The Defense Department stated that it can reconcile its missile quantity balances because it has maintained positive control of missile quantities. It also emphasizes that neither GAO nor any other investigative source has confirmed that any lapses in control have lead to theft or unexplained loss of Category I missiles. We continue to believe, however, that the Defense Department does not know how many Category I missiles of each type it should have in inventory. Therefore, it cannot be certain that a reconciliation of missile quantities includes all the missiles it should. Without such information, the possibility exists that missiles could be missing. Furthermore, the fact that we found a missile and parts of a missile missing from their containers should heighten concerns over the vulnerability of the missile inventory system. Further, item managers and some inventory managers were unable to provide accurate information regarding missile quantities and locations. The Defense Department took exception to our portrayal of its handling of an allegation regarding the theft of Stinger missiles. It said it completed an appropriate investigation and conducted a 100-percent inventory of Stinger missiles at the depot. The Defense Department could not provide details of the investigation or documentation that an inventory had been conducted at the time of the allegation. The Department's detailed comments are included in appendix II. SCOPE AND METHODOLOGY ------------------------------------------------------------ Letter :9 We inventoried Category I missiles at all Defense Department storage locations worldwide. These missiles were stored in 78 locations in the United States and at U.S. bases in Europe, Asia, and South America. Details of how we conducted our inventories are included in appendix I. We also met with Army, Marine Corps, and Navy officials involved in controlling Category I missiles to discuss each service's management process. We excluded the Air Force because it had purchased few Category I missiles. We also met with intelligence and law enforcement officials to discuss terrorist or criminal demand for Category I missiles and to identify any known diversions of the missiles from Defense Department or other inventories. ---------------------------------------------------------- Letter :9.1 Unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from its issue date. At that time, we will send copies to the Chairmen of the House and Senate Committees on Armed Services; the Secretaries of Defense, the Army, the Navy, and the Air Force; the Commandant of the Marine Corps; and other interested parties. Copies will also be made available to others on request. This report was prepared under the direction of Donna M. Heivilin, Director, Defense Management and NASA Issues, who may be reached on (202) 512-8412 if you or your staff have any questions. Major contributors to this report were Nomi R. Taslitt, William K. Newman, and Yolanda C. Elserwy. Sincerely yours, Frank C. Conahan Assistant Comptroller General SCOPE AND METHODOLOGY =========================================================== Appendix I To determine whether Category I missiles are adequately controlled and accounted for, we physically inventoried the Stinger, the Redeye, and the Dragon missiles, by serial number, at all Defense Department storage locations worldwide. Category I missiles located at sea or in other locations were inventoried, at our request, by the military services. We did not attempt to inventory the missiles that were sold or transferred to foreign governments. We inventoried all identified storage locations in the continental United States between May and October 1992. We then inventoried missiles at all identified foreign locations, including Europe, Asia, and Latin America, completing our inventory in April 1993. We completed reconciliation of the data in March 1994. Each service gave us a list showing where each type of missile was stored and how many were stored at each location. The services notified each location of our inventory plans well in advance of our visits. Before our visits, we also directly notified storage location personnel of the dates of our planned inventory. At each site, we (1) asked the storage manager to identify all locations where Stinger, Dragon, or Redeye missiles were stored; (2) inventoried all Stinger, Dragon, or Redeye missiles found in the locations identified by the storage manager; (3) hand recorded the serial number imprinted on each missile container or, if the serial number was missing or hidden, recorded the serial number found on the bar code, the pallet card, or a paper tag; (4) opened about 1 percent of the missile containers to ensure that they contained a missile and that the missile and container serial numbers matched; (5) observed the physical security provided Category I missiles at each location; and (6) asked storage managers to provide a list of the serial numbers of all the missiles stored at the facility during our visit and a list of the serial numbers of missiles received during the 3 months following our inventory. We conducted our review from May 1992 to March 1994 in accordance with generally accepted government auditing standards. (See figure in printed edition.)Appendix II COMMENTS FROM THE DEPARTMENT OF DEFENSE =========================================================== Appendix I (See figure in printed edition.) (See figure in printed edition.) Now on pp. 1-3. (See figure in printed edition.) (See figure in printed edition.) Now on pp. 4-5. (See figure in printed edition.) (See figure in printed edition.) (See figure in printed edition.) Now on pp. 5-9. (See figure in printed edition.) (See figure in printed edition.) See comment 1. Now on p. 10. (See figure in printed edition.) (See figure in printed edition.) Now on pp. 10-11. Now on p. 11. (See figure in printed edition.) Now on p. 11. (See figure in printed edition.) Now on p. 11. Now on p. 11. (See figure in printed edition.) Now on p. 11. Now on p. 12. (See figure in printed edition.) The following is GAO's comment on the Department of Defense's letter dated July 18, 1994. GAO COMMENT 1. We made two comparisons for each missile storage location inventoried. We compared the number of missiles found at each storage location with (1) the number of missiles identified on item manager records and (2) the number of missiles identified on storage site records. The two comparisons often produced very different results. Item managers told us there were 2,185 more Stinger, Dragon, and Redeye missiles at Europe's 11 storage sites than we found, and storage site records identified 4 less than we found. RELATED PRODUCTS Ammunition and Firearms Accountability: 24th Infantry Division and Fort Stewart (Army Audit Agency Report SR 92-8, Jan. 31, 1992). Special Report on the Review of Redeployment of Personnel, Equipment, and Materiel from Saudi Arabia (Army Audit Agency Report SR 92-309, Aug. 25, 1992). Controls Over Ammunition and Explosives (DOD/OIG Report 91-119, Sept. 11, 1991). Review of Ammunition Accountability (Army Audit Agency Report EC-91-711, Sept. 6, 1991). Ammunition Accountability Audit at Fort Stewart, Georgia (Internal Review Report A1-90, Feb. 4, 1991). Army Inventory: A Single Supply System Would Enhance Inventory Management and Readiness (GAO/NSIAD-90-53, Jan. 25, 1990).