SECRETARY OF LABOR, Complainant, v. INLAND STEEL COMPANY, Respondent. UNITED STEELWORKERS OF AMERICA, AND ITS LOCAL 1010, Authorized Employee Representative. OSHRC Docket No. 79-3286 _DECISION_ Before: BUCKLEY, Chairman; and RADER, Commissioner.[[*]] BY THE COMMISSION: This case is before the Occupational Safety and Health Review Commission under 29 U.S.C. § 661(j), section 12(j) of the Occupational Safety and Health Act of 1970, 29 U.S.C. §§ 651-678 ("the Act"). The Commission is an adjudicatory agency, independent of the Department of Labor and the Occupational Safety and Health Administration. It was established to resolve disputes arising out of enforcement actions brought by the Secretary of Labor under the Act and has no regulatory functions. _See_ section 10(c) of the Act, 29 U.S.C. § 659(c). This litigation was precipitated by the death of an Inland Steel Company switchman, Anthony Corona, in one of Inland's railroad yards. Corona was fatally injured when he was crushed between two groups of railcars which he was attempting to couple together. The Secretary's Occupational Safety and Health Administration subsequently inspected Inland's rail operations and issued a citation alleging that Inland willfully violated section 5(a)(1) of the Act, 29 U.S.C. § 654(a)(1),[[1]] in that it did not adequately protect its employees from being injured by moving railcars during switching or transporting operations. The specific allegations of the citation, as amended, were (1) that Inland failed to equip its railcars with handbrakes and (2) that it did not have an adequate safety program for reporting and removing from service railcars with defective safety equipment and for preventing employees from working with those cars. It was undisputed that Inland did not install handbrakes on its railcars. Rather, Inland contended that handbrakes were less effective and more dangerous than wheel blocks. For this reason, Inland actually removed handbrakes from its railcars and instead used specially designed wheel blocks to hold railcars in position. Inland also contended that its rules for identification and removal of defective railcars were adequate and were communicated to its employees. Commission Judge Richard DeBenedetto vacated both charges. He found that given the limitations and hazards of handbrakes in Inland's operations, Inland's decision to prevent "inadvertent movement of railroad cars" by using blocks was proper. The judge also found that Inland's safety rules for reporting and removing from service defective railcars were adequate and that they were properly communicated to employees. Finally, Judge DeBenedetto found that the evidence did not sustain the Secretary's allegation that Inland failed to adequately enforce the rules. We affirm Judge DeBenedetto's action in vacating the amended citation. I. _The Handbrake Allegation _ A Inland's Harbor Works in East Chicago, Indiana, is the largest steel manufacturing facility in North America, comprising 2500 acres and employing 24,000 people. Inland uses some 3000 railcars of different types in its internal railroad operations, which include 150 to 175 miles of railway track. Its cars move the materials used in its steelworks, including molten metal, steel ingots, and raw materials like scrap metal. In addition, hundreds of "foreign" railcars (originating from outside railroads) are handled within the facility each day. It is the safety of Inland's 250 or so switchmen that is the focus of this case. Switchmen couple railcars to trains and uncouple them, and perform the switching which allows cars to transfer from one track to another. Inland's switching crews consist of three persons--a "conductor," who is the head switchman, another switchman, also called the "helper," and a locomotive engineer. Inland's operations consist basically of flat switching, that is, switching railcars on level tracks. Inland's yards are "dish-shaped," that is, level except for an incline at their sides. Inland typically sorts its railcars into their appropriate track by releasing or "kicking" them down the inclined edge of the yard, letting them roll freely over switching points until they come to a standstill at their appropriate track. Most of its cars are in more or less constant use, being loaded or unloaded, with intermediate switching. Unlike foreign railcars, Inland's railcars do not have handbrakes. In 1958, Inland implemented a system to prevent movement of stationary railcars in areas where employees are working by using wood blocks shaped like parallelograms, 2" high, 2" thick, and 11" long (measured diagonally). Inland instructs employees in the use of different types of blocking for different circumstances, to preclude movement of the cars after blocks are set. Each year, Inland's employees block over 100,000 cars. The amended citation alleged a willful section 5(a)(1) violation in that: The employer did not provide hand brakes on employer owned railroad cars used in the railroad yards plant wide as set forth below: Employees were exposed to the hazard of being struck by, caught between, or run over by moving railroad cars while conducting switching and transporting operations as the railroad cars were without adequate means to be braked . . . . B The judge found no evidence that other employers in Inland's industry, the steel industry, "use handbrakes instead of blocks to secure plant railway freight cars." He also found that in the industry of interstate railroad carriers "serious risks attending the use of handbrakes persist." Specifically, employees climbing on and off railcars to set and release handbrakes must contend with a "precarious situation . . . . Railroad accident data reveal that getting on and off cars and operating handbrakes are major causes of injury. . . . [T]he Secretary has failed to show by any competent evidence that the use of blocks has been the cause of injury to Inland employees." The judge determined that wooden blocks were not shown to be unacceptable in either the steel industry or among interstate railroads for the uses to which Inland puts them. He also determined that an "elaborate inspection and maintenance program" would be required for handbrakes on Inland's 1400 or so gondolas--about half its fleet--due to the extensive damage to which they would be subjected in Inland's scrap handling operations. He found that the total repair problems might very well be an insurmountable task and that the "record fully justifies Inland's decision" to use specially designed blocks instead of handbrakes. We agree. C To prove that an employer violated section 5(a)(1), the Secretary must show that the employer failed to free its workplace of a hazard that is recognized as such by the employer or generally by the employer's industry, that is causing or is likely to cause death or serious physical harm, and that could have been materially reduced or eliminated by a feasible means of abatement. _E.g_., _Cerro_ _Metal Products Division, Marmon Group Inc_., 86 OSAHRC __/__, 12 BNA OSHC 1821, 1822, 1986 CCH OSHD ¶ 27,579, p. 35,829 (No. 78-5159, 1986). The judge in this case defined the recognized hazard as "inadvertent movement of railroad cars." The Secretary [[2]] urges us to accept that formulation.[[3]] However, Inland objects to this definition on the ground it is overly broad, contending that the hazard should be defined as the absence of handbrakes. As we have held, defining recognized hazards too broadly would undermine the congressional purpose behind the "recognition" element-- to limit the general duty imposed by section 5(a)(1) to preventable hazards. _Pelron, Inc_., 86 OSAHRC, ___, 12 BNA OSHC 1833, 1835, 1986 CCH OSHD ¶ 27,605, p. 35,871-72 (No. 82-388, 1986); _see_ _also_ _National Realty & Construction Co. v. OSHRC_, 489 F.2d 1257, 1266 (D.C. Cir. 1973) (section 5(a)(1) limited to preventable hazards). We observed in _Pelron_ that an employer cannot reasonably be expected to free its workplace of inherent risks that are incident to its normal operation. Therefore, to respect Congress's intent, hazards must, be defined in a way that apprises the employer of its obligations under the Act by identifying conditions or practices over which the employer can reasonably be expected to exercise control. 12 BNA OSHC at 1835, 1986 CCH OSHD at p. 35,872, _citing_ _Davey Tree Expert Co_., 84 OSAHRC 11/D11, 11 BNA OSHC 1898, 1984 CCH OSHD ¶ 26,852 (No. 77-2350, 1984). Consistent with this precedent, Inland argues that the hazard in question must be defined as a lack of handbrakes in order to afford it fair notice of the conduct the Secretary is requiring. Since movement of railcars is inherent and unavoidable in Inland's operations, Inland is not required under section 5(a)(1) to free its worksite of all railcar movement that might present a potential risk to its employees. Here, Inland had instituted a program for the use of wheel blocks, the purpose of which was to prevent undesired or unintended movement, or movement beyond that necessary for the conduct of its usual operations. However, while we conclude that "inadvertent" is an appropriate term to characterize the type of railcar movement from which Inland sought to protect its employees, we need not decide whether the recognized hazard may be defined in this manner or must be defined more narrowly as Inland contends. In a case such as this, where the employer has a mechanism designed to eliminate a hazardous condition, the burden is on the Secretary to establish that the employer's measures were inadequate. _Cerro Metal Products_, _supra_, 12 BNA OSHC at 1822-23, 1986 CCH OSHD at p. 35,829, _citing_ _National Realty & Construction Co. v. OSHRC_, 489 F.2d 1257, 1266-68 & n.40 (D.C. Cir. 1973). In addition, the Secretary must show that knowledgeable persons familiar with the industry would regard additional measures as necessary and appropriate in the particular circumstances existing at the employer's worksite. _Id_. We conclude that the Secretary has failed to meet that burden in this case. Since the Secretary has not established that Inland should have implemented the abatement measure he advocates--the use of handbrakes---the citation allegation must be vacated regardless of how the recognized hazard in issue is defined. _See_ _Pelron_, _supra_, 12 BNA OSHC at 1835, 1986 CCH OSHD at p. 35,871. (1) Inland is a member of the steel industry. Many steel plants contain specialized railroads similar to Inland's, but the Secretary submitted no evidence to show steel industry practice or knowledge. In fact, Harding, superintendent of Inland's Transportation Department, testified that steel plants normally do not have handbrakes on their railcars. Thus, the evidence in this case does not indicate general understanding in the steel industry that blocks are not an adequate means to protect employees from inadvertent railcar movement.[[4]] _See_ _Pratt & Whitney Aircraft v. Secretary of Labor_, 649 F.2d 96, 101 (2d Cir. 1981). (2) It is conceded that Inland is not engaged in business as a common carrier by railroad. Despite this, the Secretary argues that the interstate railroad industry is a relevant industry, that it recognizes the need for handbrakes in operations comparable to Inland's, and that Inland is responsible under section 5(a)(1) for following the same practices. We disagree. First, the Secretary did not prove that interstate common carriers generally use handbrakes, in their flat switching operations. Second, those operations are not comparable to Inland's. Inland's switching differs markedly because it is much more frequent, increasing greatly the serious risks of handbrake operation, and because handbrakes would be subject to much greater damage than in switching operations of interstate common carriers. In addition, Inland regularly uses types of railcars, such as ingot buggies, double pot cars and pan cars, that are found rarely if ever among interstate carriers. As described below, these cars will not accept a standard handbrake and carry extremely hot or molten metal, creating special problems for switchmen attempting to secure them from movement. The Secretary relies on the fact that railcars used in interstate commerce must be equipped with efficient handbrakes under the Safety Appliance Act, 45 U.S.C. § 11. However, there is no explicit statutory requirement that interstate common carriers use handbrakes either generally or in switching operations similar to those involved in this case. In attempting to prove that interstate railroads _use_ handbrakes, the Secretary relies on the testimony of Marshall Sage, a rail safety specialist with the United Transportation Union, and Wilder Barnes, a former Federal Railroad Administration safety specialist. They testified that the safety rules of interstate carriers generally require that all railcars be properly secured with handbrakes when left on any track. However, many of the railroads' rules merely specify that a "sufficient" number of brakes be applied to prevent cars from moving. Sage admitted that the term "sufficient" is not clear and has been interpreted in the railroad industry to mean that brakes do not necessarily have to be used in all instances when cars are left on tracks.[[5]] Barnes also admitted that blocks are used in railroad switching operations for temporarily holding cars, and Sage stated that some interstate carriers' rules recognize that brakes may not always be adequate to prevent movement and require blocks to be used in such instances. Holland, a former Inland employee who had prior rail carrier experience, stated that he had never used handbrakes except to hold cars on elevated tracks. James Martin, Senior Vice President for Operations of the Illinois Central Gulf Railway Co. ("ICG R.Co."), testified for Inland that he did not interpret the common carriers' rules to require the use of handbrakes in all cases, but merely to require that cars be secured by any sufficient means. This could mean chocks or blocks, skates, handbrakes, or nothing at all, depending on conditions. Furthermore, Martin testified without rebuttal that in most flat switching operations handbrakes are not used even by interstate carriers. Thus, the Secretary failed to show that handbrakes are generally used in the type of switching operations at issue here. Even if interstate rail carriers normally do use handbrakes in their flat switching operations, there are significant differences between the typical flat switching operations of interstate common carriers and those at Inland. The most important difference is the frequency of Inland's switching. Martin testified that Inland switches its railcars much more often than interstate railroads because they are being loaded and unloaded in materials-handling almost constantly. Harding testified that Inland switches a railcar perhaps as much as 15 to 20 times more often than outside railroads he has observed. Thus, if handbrakes were generally required, switchmen would have to climb on railcars to set or release them perhaps 15 to 20 times more often than on interstate common carriers. As a result of the greater frequency of Inland's switching, the risk of injury to Inland's switchmen while applying handbrakes would be even greater than the risk that switchmen for interstate common carriers face. On interstate rail carriers during 1975-78, approximately 6.5 percent of all injuries to yardmen (in which category switchmen would fall) involved operating handbrakes. Even if injuries on moving railcars are excluded, the Federal Railroad Administration Accident/Incident Bulletin covering 1978, introduced in evidence, indicates as many as 125 injuries to employees while operating handbrakes in switching and terminal operations, and as many as 326 other injuries while getting on or off railcars in those operations.[[6]] On the other hand, despite the large number of blocking operations performed each year, Inland's present blocking system had resulted in no injuries to employees due to blocking railcars or being struck by railcars that had been blocked during the entire 22 years the system had been in operation. As Martin explained, blocks can be placed from the ground at all times, with the employee's body clear of the car. Martin testified, "most accident prevention people would, I'm sure, opt to keep the man off the car because of the potential hazard in boarding and detraining." Martin was well qualified to give factual opinion testimony. His duties with ICG R.Co. included responsibility for safety. Martin had served on a railroad safety advisory committee for the Office of Technology Assessment of the United States Congress. The committee prepared an extensive study of the causes of railroad accidents and Federal regulation of the subject. Martin also had responsibilities for development of safe practices of Class I (major) rail carriers as Vice President of Operations and Maintenance for the Association of American Railroads. Martin's conclusion is also consistent with the recommendation of Inland's committee that formulated Inland's current blocking system in 1958. None of the five switchmen who served on that committee, most of whom were experienced in the operations of interstate rail carriers, advocated using handbrakes. The other member of the committee was then transportation superintendent, Smolt, who previously had worked as a switchman. He testified that operating handbrakes is the most hazardous part of a switchman's job. The much greater damage handbrakes would be exposed to in Inland's operations also makes any comparison between Inland and interstate carriers inapposite. The judge found that if handbrakes were required on Inland's 1400 or so gondolas--about half its fleet--damage to the handbrakes would be so frequent that the repair problems "may very well be an insurmountable task." Martin noted that in Inland's scrap metal transportation operations "there is no way to keep safety appliances in working order," due to the almost constant damage to which the gondolas are exposed from huge pieces of scrap metal-- individual pieces weighing as much as 4 or 5 tons--being loaded into and unloaded from the cars with electromagnets. Martin explained that different considerations apply with respect to interstate carriers. Although there is a limited amount of scrap handling done on interstate rail carriers, there is not the frequent exposure of safety appliances to heavy damage due to constant loading and unloading. Instead, interstate carriers use switching yards primarily to receive, sort out, and dispatch trains originating from, and destined for, other points. Inland's frequent transportation of extremely hot or molten metal on specially designed cars is another important difference between its operations and those of interstate common carriers. Many of Inland's cars are regularly involved in the transport of extremely hot or molten metals. These include 500 or so ingot buggies, 50 double pot cars, 31 or more pugh ladle cars, and an unspecified number of pan cars. Martin testified that ingot buggies, double pot cars and pan cars would not accept a standard braking system, so a special system would have to be devised for them. The extreme heat of the metals carried on many of Inland's cars would preclude mounting the car to set handbrakes, whereas blocks may be set because the switchman is shielded somewhat from the heat by the car's body when setting them. Martin noted that interstate common carriers rarely carry hot metal. Thus, numerous important differences in working conditions and hazards make the Secretary's comparison between the braking practices of interstate rail carriers and Inland inappropriate. The interstate railroad industry simply does not furnish a model that Inland should be expected to follow. _See_ _Donovan v. Royal Logging Co_., 645 F.2d 822, 830 (9th Cir. 1981) (practice of one industry may not be imposed on another industry where hazards and working conditions are not comparable). (3) The evidence also fails to show that knowledgeable persons familiar with Inland's industry would regard Inland's system for the use of blocks to be ineffective in protecting employees. As noted above, Inland had great success in safely holding railcars with blocks, and it had no injuries attributable to its present blocking system in the 22-year history of that system. Numerous witnesses experienced with both common carriers and in-plant railroads testified that Inland's blocking system was quite adequate, even when compared to handbrakes. Accordingly, the Secretary's argument that the use of blocks rather than handbrakes would increase the likelihood of railcar movement in certain specific situations is not supported by the record. Transportation superintendent Harding testified that properly blocked cars will hold railcars in almost all situations, including on icy or wet rails. Inland employees are advised of different types of blocking to be used in different circumstances and are instructed to block all cars so that there will be no movement after they are set. Inland also has procedures to test the sufficiency of the blocking before the railcars are left on any track. Harding testified that where the incline is away from the engine, the initial blocking should be tested before the blocked cars are uncoupled from the train, by backing the train slightly to see if the blocks hold. If they do, there is reasonable assurance that the blocking is sufficient, and the railcars are then uncoupled from the train. Where the incline is toward the engine, the train may be pulled slightly forward to test the blocks, or the blocked cars may be uncoupled and the engine may pull away a short distance. In the latter case, if the blocks fail the blocked cars will roll toward the engine and couple up again, and additional blocking will be used. Although the process of setting blocks sometimes involves more than one attempt, switchman Thedford, a former union safety steward, testified that if an initial blocking failed, he would simply try other blocking methods until a successful combination was found. Thus, Inland's procedures were designed to minimize the chance of railcars rolling free after they were blocked. Martin testified that Inland's blocking system is "certainly very adequate" for securing railcars in its operations, "very safe" and actually "superior to the handbrake . . . in the type of activity they are involved in." Martin had visited Inland's facility and had observed its blocking system in use. His opinion was also based on his observation of other intraplant railroad facilities in the steel industry that used blocks rather than handbrakes, including Bethlehem Steel Company, Armco Steel Corporation, and Youngstown Sheet and Tube Company. William Webber, an Inland trainmaster, and Charles C. Shannon, an independent railroad consultant, also testified that Inland's blocking system was fully adequate for securing its railcars. They also were well qualified to give such an opinion. Webber previously had been a switchman for Inland for more than five years, and before that had been a brakeman, operating handbrakes for the Pennsylvania Railroad for about five years. Shannon, an independent railroad consultant since 1960, had previously been a trainmaster qualified in and responsible for instructing switchmen on railroad safety rules including handbrake use for the Chicago and Northwestern Railway and had a total of 19 years managerial experience with that railroad. While employed by the railroad, Shannon personally had set blocks and applied brakes. Like Martin, he also observed Inland's blocking system first-hand. The Secretary relies on evidence that blocks could fail on Inland's pugh ladle cars (molten metal carriers) due to the great weight of the cars, water and mud on the tracks, the incline of the track; and the swaying of the cars when the molten metal is poured into them. However, Harding testified that pugh ladle cars generally hold during loading operations when properly blocked on both sides of a wheel. The only way that a properly blocked pugh ladle could move in Harding's opinion was if molten metal spilled and burned the blocks or if some equipment such as a bulldozer struck the car--both very rare occurrences, based on his testimony. Furthermore, Harding testified that in his opinion a handbrake would not hold a loaded pugh ladle either. Webber testified that the high temperature of the hot steel carried on ingot mold buggies and stake cars would preclude the use of handbrakes, whereas those cars could be blocked because temperatures are cooler near track level. The Secretary further argues that handbrakes were needed because railcars generally could be struck by machines during loading or unloading and roll free of their blocks. While arguably further braking measures might decrease the chance of movement in that limited situation, there is no evidence that any employee had ever been endangered by such an occurrence nor any evidence from which to determine the likelihood or the frequency of such occurrences. Barnes, a witness for the Secretary, pointed out that handbrakes have the holding power of 11 percent of the gross car weight; that they apply equally to all eight wheels on the typical railcar; and that they retain some braking power even if the railcar is set in motion, whereas blocks have none once a wheel rolls over them. For example, Barnes testified that a railcar could be derailed if Inland's strongest blocking method, crossblocking--where two blocks are placed at right angles so that one braces the other--were used on one set of wheels and the crossblocked car were struck or pushed. There is no evidence that such a derailment had ever actually occurred. Barnes was the only witness to testify to this possibility, and he was not asked to describe the likelihood of such occurrences in light of the conditions existing in Inland's facility. Barnes also testified that in his opinion Inland's blocking procedure was hazardous because of the proximity of the employee's hand to the wheel in setting the block and the amount of time required in those situations where more than one block must be set. However, Barnes had never visited Inland's facility. There is no indication that he had any knowledge of operations of internal railways in the steel industry, nor was he familiar with the type of blocks and the details of the blocking system instituted by Inland. The committee which developed Inland's present blocking system had established based on actual tests that the special, tapered design of Inland's blocks eliminated the hazards of employees' fingers or hands being pinched between a block and a rail, or between a block and a wheel, when placed properly. The judge was not persuaded that Barnes' testimony, or that of Sage, outweighed the evidence supporting Inland's position. We cannot say that the judge erred.[[7]] The Secretary presented testimony that blocks had been dislodged if they were "green" (insufficiently cured), or if the blocks or rails were wet or icy. However, Transportation Department Superintendent Harding testified without contradiction that cars would not move even under those circumstances if the blocks were properly set as employees were taught and that properly placed blocks would hold railcars in almost all situations. The number of such incidents of blocks failing to hold were few and, as shown below, Inland closely supervised compliance with its blocking procedures and disciplined employees who failed to properly block. There is other employee testimony, on which the Secretary also relies, that at times inadvertent movement of blocked cars had occurred because a car had been kicked into a track and had hit a blocked car, causing it to roll free of its blocks. However, many of Inland's tracks have permanent derails and/or locking switch stands. These devices prevent cars from moving into the protected track. Also, Harding testified that when employees are working on or near a track, they will place portable derails on both ends of the track or at one end if it is a stub track. There was no evidence from which to conclude that these precautions were not adequate to protect employees from being struck by a car dislodged from its blocks by a kicked car. The adequacy of Inland's blocking system is also objectively demonstrated by the relatively low incidence of block failure compared to the total number of blocking operations performed. Thus, Thedford stated that he had experienced blocks failing to hold a car only once or twice a week. Employee Warner testified that during the six months he had worked as a switchman he had blocked cars on numerous occasions and only in 10 instances did the blocks fail to hold. Although the details of Inland's safety program are discussed more fully in the Part II, we note that Inland monitors compliance with its safety rules and disciplines employees for infractions. During the period from 1974 through 1979, Inland reprimanded or penalized employees on 25 occasions for failure to block cars properly. Section 5(a)(1) does not require an employer necessarily to prevent every instance of hazardous conduct by its employees. _See_ _Jones & Laughlin Steel Corp._, 82 OSAHRC 34/A2, 10 BNA OSHC 1778, 1782-83, 1982 CCH OSHD ¶ 26,128 at pp. 32,887-88 (No. 76-2636, 1982). The Secretary has abandoned his claim that Inland should require the use of handbrakes to slow or stop moving railcars.[[8]] However, the union has not expressly abandoned that argument. Leaving aside the question of whether the union may press claims that the Secretary has abandoned, we find Inland's procedures for handling moving cars to be adequate to protect its employees. We agree with the judge that the speed of the cars is adequately controlled by the locomotives, which move at a slow rate of speed, and by the basically flat topography of dish-shaped switching yards. In fact, as the judge found, "[t]he evidence indicates that having switchmen dismount the car and safeguard road crossings in the path of the 'runaway' car, as Inland does, is a safer practice in dealing with the situation than the Secretary's recommended method of requiring switchmen to manipulate hand brakes of a moving car." In sum, none of the evidence relied on by the Secretary shows a need for Inland to use brakes to protect employees from railcar movement. Because the Secretary failed to establish that Inland did not take sufficient steps through its blocking system and instructions to employees to keep its workplace free of the hazard of inadvertent movement of railcars, he has failed to prove a section 5(a)(1) violation as to this item. _United States Steel Corp_., 81 OSAHRC 28/C2, 9 BNA OSHC 1641, 1981 CCH OSHD ¶ 25,282 (No. 76-5007, 1981). _See_ _Brennan v. OSHRC (Republic Creosoting Co._), 501 F.2d 1196, 1200 (7th Cir. 1974). II. _The Safety Program Allegation_ The remaining issue on review is whether the judge erred in ruling that the employer's safety program was adequate. The Secretary's amended citation alleged three specific deficiencies in the safety program as follows: A. The employer did not maintain railroad cars in a _safe_ operating condition. The employer's use of defective railroad cars exposed employees to the hazard of being struck by, caught between, or falling from such railroad cars while switching or transporting the cars. Railroad cars with defective or missing uncoupler devices (pin lifters), bent and crushed stirrups, bent, crushed, loose and missing handholds (grab irons) and broken yokes (carrier irons) were being used rather than taken out of service as set forth below: [list of individual railroad cars omitted] B. The employer failed to enforce company safety rules requiring transportation employees to identify and report defective railroad cars having missing or defective uncoupler devices (pin lifters), bent or crushed stirrups, bent, crushed, loose or missing hand holds (grab irons) and broken yokes (carrier irons). C. Switchmen and Conductors working in railroad yards plantwide are required and/or permitted to work with defective railroad cars having defective and/or missing uncoupler devices (pin lifters) necessitating stepping between cars to open couplers manually.[[9]] (Emphasis in original) It is undisputed that at times some of Inland's railcars in service had defective pin lifters, stirrups or handholds; that switchmen did not always report defective railcars; and that at times some switchmen went between railcars to uncouple cars manually, due to missing or defective pin lifters. It also is undisputed that Inland had an extensive safety program including rules designed to prevent these problems, communication of the rules to switchmen, and efforts to detect violations. The basic issues in dispute are whether Inland's safety procedures for reporting defective railcars and uncoupling railcars were adequate, and whether its safety rules were adequately communicated to employees and adequately enforced. The judge found that Inland had established an adequate safety program and that its employees were adequately trained in the pertinent safety rules. He rejected as not credible certain testimony introduced by the Secretary to show that supervisors coerced or induced employees to violate these rules. He also found that the Secretary did not provide a sufficiently definite standard or criterion against which to judge Inland's maintenance program and that the Secretary failed to show what additional measures Inland could have taken to insure compliance with its safety rules. He determined that the Secretary's evidence regarding feasible abatement methods lacked probative value because the witness on whom the Secretary relied, the compliance officer, was not familiar with the circumstances of either the steel industry or of railroads in general. As noted above, to establish a section 5(a)(1) violation, the Secretary must prove, among other things, the existence of a feasible and useful means of reducing or eliminating the alleged recognized hazards. The Secretary argues that Inland's safety rules for reporting and removing defective railcars from service and uncoupling railcars were inadequate, and that the communication and enforcement of those rules was deficient. Under Commission precedent, when elimination of a recognized hazard requires that employees follow safe procedures, an employer is not in violation of section 5(a)(1) if it has established workrules designed to prevent the hazards from occurring, has adequately communicated the workrules to the employees, has taken steps to discover noncompliance with the rules, and has effectively enforced the rules in the event of noncompliance. _Jones & Laughlin Steel Corp_., _supra_, 10 BNA OSHC at 1782, 1982 CCH OSHD at p. 32,887. We find that the Secretary failed to prove the alleged inadequacies in Inland's safety program. We find specifically that Inland had adequate safety rules requiring the reporting and removal from service of railcars with defective safety appliances, and prohibiting switchmen from going between railcars where hazards of railcar movement existed; that those rules were properly communicated to employees; and that the Secretary's evidence of inadequate enforcement of these rules is insufficient.[[10]] We will address these findings in order. A Inland's safety rules directly address and would eliminate the problems complained of by the Secretary and Union. Inland's Rule 4 addresses the reporting and removal from service of defective railcars. The term "bad order" as used in the rule means that the railcar has a defective safety appliance or some other defect warranting removal from service. The rule states: Conductors and switchmen must report all defective equipment to their Yardmaster. Bad order foreign empty cars are to be set outbound as soon as possible. Bad order foreign loaded (inbound) cars are to be set for unloading as soon as possible and then set outbound empty immediately when released. Exception--Foreign cars with broken drawbars are to be set aside in the plant for "I.H.B." or "J" to repair. Red "bad order" cars signify major defects. A car with a red card must not be moved unless authorized by the Trainmaster or Asst. Superintendent and then it must be handled carefully to specific destination. [[11]] In addition to requiring switchmen to report defective equipment, Inland's Transportation Department has a field repair crew consisting of two supervisors and about 23 car repairmen assigned to inspect, repair and oil railcars in the yards daily. Yard clerks also are supposed to locate and report all defective equipment in the course of their duties. There is no evidence that switchmen, yardmasters, or car repairmen could not determine when a railcar is defective. So far as this record shows, Inland has adequate procedures for reporting defective railcars and removing them from service. As to the problem of switchmen working between railcars, Inland's Rule 43 provides: Before stepping between cars to open knuckles, adjust drawbars and etc., the cars must be at a complete standstill and a 20-foot-opening between the cars. Also if there is any question of the cars not being at a complete standstill they must be blocked. At night when it is necessary to step between cars to work on knuckles, drawbars, etc., switchman must place his lantern on the ground where it can be seen by the other switchmen or engineer. Engineer, is not to move the cars while the lantern is on the ground. In 1977, Inland further restricted work between cars by issuing a new Rule 60 (later renumbered as Rule 58 in Inland's revised rulebook). That rule provides: Under no circumstances will switchmen be permitted to lift pins by hand without using cut levers. 1. When cut lever is missing or inoperative switchman must cross in a safe manner to the other side of train and pull the cut lever on the other car. 2. If cut levers are missing or inoperative on both sides of the train the cars must be set out as bad orders for car repair to replace or repair cut levers. So far as this record shows, these rules would minimize the possibility of switchmen being injured by a moving railcar while working with defective equipment. B We also find that the safety rules were properly communicated to switchmen. Inland had an extensive safely program for switchmen that included instructions on the safety rules. Switchmen were given three days of classroom training and seven days of on-the-job training before starting work. Included in this training were instructions on the pertinent safety rules and on safe procedures for the specific jobs switchmen must perform, including blocking cars, pulling pins and coupling cars. The latter training is based on numerous detailed job safety analyses ("JSAs") that Inland has developed for particular aspects of the switchman's job. A test is administered to the switchmen at the end of the 10-day course. Following the course, each switchman-trainee spends two weeks as an extra member of switch crews. Three different conductors must sign a card saying that the trainee works safely and can handle the job. During the following three months each trainee is placed on a particular job for a week at a time with an experienced conductor. The evidence establishes that the switchmen knew Inland's safety rules. Switchman Thedford testified that each day a safety rule pertaining to their jobs is read to the switchmen, and he remembered that Rule 43 had been read occasionally. Dills, a switchman who had previously been a foreman, had learned the rules from instruction by yardmasters. Warner, formerly a switchman, had been trained in safety rules, including Rules 4, 43, and 60, and had been specifically instructed not to uncouple cars lacking pin lifters. Former switchman Connelly was familiar with Rules 4 and 43, and he indicated that he had learned of the rules through safety bulletins issued and posted by Inland. The Secretary relies on conductor Gillie's testimony that he did not know about Rule 60 until after the citations were issued and that he had not seen it posted. However, Transportation Superintendent Harding testified that Safety notices--the form in which Rule 60 was first issued--are posted in the different yard areas for the switchmen to read, and are the subject of safety contacts and observations. The judge found Gillie's testimony, including that claiming ignorance of Rule 60, not to be credible. The principles governing our determination of the correctness of a judge's credibility finding are well established. Normally, we will accept the administrative law judge's evaluation of the credibility of witnesses because it is the judge who has lived with the case, heard the witnesses and observed their demeanor. _C. Kaufman_, 78 OSAHRC 3/C1, 6 BNA OSHC 1295, 1977-78 CCH OSHD ¶ 22,481 (No. 14249, 1978). However, the judge should identify the conflicting testimony and explain the reasons for failing to credit a witness's testimony or for crediting the testimony of a witness over that of another. _P & Z Co_., 77 OSAHRC 211/F5, 6 BNA OSHC 1189, 1192, 1977-78 CCH OSHD ¶ 22,413, p. 27,024 (No. 76-5007, 1981) (judge's credibility finding will be upheld where opposing party does not demonstrate reasons why it should be reversed and Commission's review of record indicates there is no basis for doing so). We find upon review of the entire record that the judge evaluated the voluminous evidence and conflicting testimony carefully and impartially. He also specifically identified much of the testimony by Gillie that he failed to credit, and stated the reasons: Mr. Gillie's testimony is strikingly unconvincing. Time and time again he displayed bias against Inland, particularly in his demeanor and the manner in which he responded to questions on cross-examination. The major points of incredibility in his testimony include the alleged incidents of being harassed by Inland's supervisors if he failed to undertake the dangerous practice of lifting pins by hand . . . the alleged lack of knowledge concerning Inland's Safety Rule No. 60 which prohibits that very practice . . . the amount of defective cars alleged to be as much as 70 percent and his claimed efforts to alleviate the problem . . . . It is significant to note that although he had held the responsible jobs of "union griever" and "safety committeeman," neither Mr. Gillie nor any other conductor or switchman has asserted his rights under Article 14, section 6, of the collective bargaining agreement concerning defective cars [permitting an employee to be relieved of work that he believes is unnecessarily hazardous]. . . . It is also noteworthy that despite his professed concern regarding defective equipment and its hazards, Mr. Gillie failed to take any action on lnland's offer of establishing a joint union-management committee to investigate the problem of bad order cars, which was raised by Mr. Gillie himself in an October 1978 "Safety Agenda" while Mr. Gillie was acting as a safety committeeman . . . . (citations to the record omitted) The judge's decision not to credit Gillie's testimony, including his assertion that he had not known Rule 60, is sufficiently explained on the basis of factors that are peculiarly observable by the judge: Gillie's demeanor on the witness stand and his manner of responding to questions on cross-examination. The Judge perceived a bias against Inland that made Gillie's testimony unreliable. We cannot say that this well-explained finding is erroneous. We also note that Gillie's testimony on Rule 60 conflicts with Superintendent Harding's testimony that safety notices such as Rule 60 are posted for switchmen to read. We accept and concur with both the judge's credibility finding regarding Gillie and his finding that the safety rules were adequately communicated to the employees. C The remaining issue is whether, despite Inland's extensive safety program, the Secretary proved that the safety rules were not properly enforced. The Secretary argues that switchmen routinely breached the rules, that Inland knew or should have known of the employee violations, and that supervisors participated in and actively encouraged safety rule violations. The evidence supports the Secretary's assertion that at times some switchmen violated the rule against standing between cars to operate defective uncoupling devices manually. For example, Warner testified that he did so one or more times per and Connelly testified that he did so three or more times per shift, even after Inland's Rule 60 was implemented in 1977. Both Thedford and Connelly testified that the requirement in Rule 43 that switchmen never work between railcars that are less than 20 feet apart is sometimes violated because couplers will not connect properly when cars are brought together from that distance, particularly on curved track. However, the mere fact that some switchmen violated safety rules at times on Inland's vast worksite does not prove a section 5(a)(1) violation by Inland. Inland employs some 250 switchmen, who perform innumerable switching operations. Also, switch crew members often had to work out of each other's sight and out of sight of their supervisors, the yardmasters, who are in a central office or tower in each yard and cannot observe individual switchmen at work in most instances. For example, regarding the Corona fatality, the yardmaster's tower in that yard was at least 100 yards from the site of the incident, too far to permit personal observation of Corona's activities. Nevertheless, Inland has programs for enforcing safety rules through monitoring employee performance and disciplining employees for violations. To monitor employee performance, Inland's APF (Accident Prevention Fundamentals) Program calls for safety contacts and observations. Safety contacts are either group meetings or one-on-one reviews between employees and supervisors on safety rules or JSAs. Observations are planned visits by supervisors to determine whether an employee actually is complying with safety rules and JSAs. Employee performance is graded, with an unsatisfactory performance indicating a need for reinstruction. If the employee is following the rules, a discussion may not be needed and may not be held. According to Lundie, Inland's safety director, the APF Program has been successful in reducing the number of accidents and injuries at Inland since it was implemented in 1965. Also, Inland's management disciplines employees when it discovers violations of safe practices. In addition to the reprimands or discipline issued for failure to block cars properly, previously referred to, Switchman Gaglio was suspended for three days without pay in 1978 for violating Rule 43. Switchman Corona was verbally reprimanded by trainmaster Webber three days before his death for violating Rule 43 and was told he would be suspended. Later, after studying Corona's personnel records and finding no previous incidents, Inland determined that a written reprimand would be issued instead. Issuance was pending when Corona died. On these facts the Secretary must show more than occasional instances of noncompliance with safety rules in Inland's yards to establish that it violated section 5(a)(1). _See_ _Jones & Laughlin_, _supra_. 10 BNA OSHC at 1783, 1982 CCH OSHD at p. 32,888. Furthermore, it is not clear on the record that employees were exposed to the hazard of inadvertent movement every time they went between cars in violation of Inland's safety rules. For example, Connelly testified that when uncoupling cars having defective lifters, he would inform the engineer and instruct the engineer not to move the engine until the uncoupling was completed. The Secretary argues, however, that yardmasters knew that switchmen violated safety rules and condoned, encouraged, and even participated in violations. The Secretary's specific arguments are that yardmasters knew that employees did consistently report defective cars and sometimes failed to ensure that bad order cars that were reported were removed from service as required by Rule 4, that yardmasters also knew that switchmen sometimes violated Rules 43 and 60, that they actually required switchmen to violate these rules by "harassing" them if they did not do so, and that yardmasters personally violated those at times. The Secretary relies on the compliance officer's testimony that yardmaster Westerfield told him that about 90 percent of Inland employees did not report defective cars, and that Harding also told him some employees did not report them. However, since Inland placed primary reliance for detecting and correcting defective railcars on its field repair crew in addition to its yard clerks who were also responsible for reporting defective railcars, Inland had procedures to make sure that defective cars were removed from service even if switchmen disobeyed the rule to report them. In any event, several switchmen testified that they did in fact report defective cars.[[12]] These employees, though, claimed that their reports were not acted upon. We do not agree. As the judge correctly observed, switchman Warner, who claimed that bad order cars he had reported were kept in service in two instances, conceded that one loaded car was first unloaded in conformity with Rule 4 and that other defective cars were simply separated from "good cars." Former switchman Connelly testified that there was a chronic failure among yardmasters to write up orders on bad order equipment that had been reported, thus preventing the equipment from being taken out of service. However, as the judge found, the only specific incident Connelly recounted in that regard "seriously undermines the import of Mr. Connelly's statements." Connelly testified that once between 1972 and 1976, yardmaster Lambert (since retired) asked Connelly to pick up four or five railcars to use in mill operations. Connelly initially objected because he considered them to be "bad order." Lambert asked Connelly to reconsider because the cars were needed at the mill, and assured him that the cars would be repaired as soon as that operation was completed. Connelly agreed. In his testimony, Connelly did not specify what the defects had been. Under Rule 4, the yardmaster must exercise judgment to determine whether a defect is serious, warranting immediate removal from service, or minor, not requiring immediate action. Thus, Connelly's account of hit agreement with Lambert actually undermines his general allegations of yardmasters failing to comply with Rule 4. Indeed, Connelly admitted that it may not always be possible to take a bad order car immediately out of service. We are not persuaded that either Warner or Connelly knew of any failure of a yardmaster to write up appropriate orders on any car with a defective or missing pin lifter or other defect warranting removal from service.[[13]] Switchman Thedford testified that he had reported bad order cars in writing to Assistant Transportation Superintendent DuVall but admitted he could not testify that those same cars were not repaired. He also testified that on one occasion he had reported to Duvall the car numbers of 20 to 25 bad order cars, and that most of the "major" repairs were not made, although some minor repairs were made. However, this testimony does not establish that DuVall did not properly report those cars to car repair and remove them from service. In fact, it suggests that he did. Thus, Thedford's testimony does not support the Secretary's position regarding the implementation of Inland's safety rules. Yardmaster Westerfield testified that he always wrote up switch orders on bad order cars and always reported railcars to car repair when he saw missing pin lifters, crushed or broken handholds, bent stirrups or cracked yokes. On an average workshift he reported up to five or ten railcars as bad order. As discussed below, the judge refused to credit other testimony that implicated yardmasters in safety rule violations and that conflicted with Westerfield's testimony. In fact, Superintendent Harding testified that switchmen do not even need a yardmaster's permission to set bad order cars to the repair tracks. We are therefore not convinced that yardmasters interfered with the removal of bad order cars from service as alleged. The employees' testimony must also be viewed in light of the evidence regarding the extent of bad orders cars. The Secretary argues that such a high percentage of railcars in service were defective that Inland's procedures for reporting and removing such cars from service must have been inadequate under section 5(a)(1). However, it was not established that percentage of cars in service was defective. Former switchman Connelly testified at one point that as much as 70 percent of Inland's cars in service were in bad order condition, but later he gave an estimate (400 bad order cars out of a total of 5000) that amounts to only about 8 percent. Gillie also testified to the 70 percent figure but the judge ruled his testimony not credible and we accept that finding for the reasons discussed above. Harding denied that Inland had found as many as one-third of cars in service to be defective. Compliance officer Gigli testified that of the 55 cars he personally examined, approximately 60 percent were in bad order condition. He could not recall the total number of cars in the yard during his inspection. Since he had no prior experience with railroad rolling stock, Gigli did not know the relative number of bad order cars that would normally be present in a typical railroad yard at any given time; his opinion that there were too many defective cars in service was based on his overall experience in an industrial environment and did not take into account the particular materials handled by Inland's cars. At the same time, however, Gigli conceded that the repair tracks were filled with cars waiting to be repaired. This evidence does not establish that Inland failed to take adequate steps to remove bad order cars from service,[[14]] nor does it show what additional measures Inland could have implemented to protect switchmen from having to work with defective gondolas or other cars. The Secretary also relies on employee testimony that yardmasters knowingly allowed or required switchmen to violate Rules 43 and 60 on occasion. However, the judge refused to credit that testimony, and we accept the judge's credibility findings, for the reasons discussed below. The other testimony also fails to show supervisory knowledge of, or involvement in, safety rule violations. First we will discuss the testimony that the judge refused to credit. Switchman Warner testified that although he was taught the correct methods of uncoupling cars by yardmaster Murray, Murray had given him advice "off the record" during that training on how to "save time and trouble" when uncoupling cars with defective pin lifters by pushing the pin up with a wooden block while standing between the cars. As the judge observed, Murray flatly denied having given any such instruction. Furthermore, Warner admitted that he would have been reprimanded for violating a safety rule if he were injured while uncoupling cars in the manner Murray suggested. In effect, Warner was testifying that a supervisor would give an instruction directly contrary to a safety rule known to be enforced by Inland. In our view, the judge could reasonably find such testimony not to be credible. Thedford testified that in early 1979 in Plant 1, yardmaster Westerfield was present when Thedford operated couplers manually because they were missing pin lifters. He also said that Westerfield and certain other yardmasters had manually opened "knuckles" for lack of pin lifters and had gone between cars that were spaced less than 20 feet apart. However, Westerfield testified that he expected employees to adhere to "serious" safety rules like Rule 43 and that disciplinary action is "almost automatic" for infractions of such rules. He testified that he had never seen, not had ever been informed of, individuals uncoupling a car that was missing a pin lifter and that he would not condone that practice. Thus, he flatly denied any implication that he was involved in safety rule violations. The Secretary also relies on the testimony of former switchman Holland that company officials knew of the practice of employees going between railcars to uncouple them when pin lifters were missing. However, Holland was retired and had not worked for Inland since 1975, before Inland's Rule 60 specifically prohibited that practice. Rule 43, the only pertinent safety rule when Holland was an employee, did not forbid it. Furthermore, Holland stated that at least on some occasions when it was necessary to operate a defective coupler, Inland's mobile car repair facility would send a repairman to install a pin lifter. Holland's testimony does not show that Inland supervisors condoned violations of any safety rules, either before or after he retired. Warner, Thedford, and Gillie made general statements that they would be subjected to "harassment" by yardmasters if they did not go between cars and manually operate coupling devices at times to keep the trains moving. They alleged that the "harassment" would take the form of being "chewed out" by the yardmaster, being restricted to the official lunch period of 20 minutes, not being allowed the routine one-hour early quitting time, or being assigned the least desirable jobs. However, of the two yardmasters who testified, Yardmaster Westerfield stated that employee violations of safety rules are not tolerated and yardmaster Murray flatly stated that the employee testimony against him was not true. Murray also stated that when training employees he always explains that cars having defective pin lifters are to be set aside as bad order and are not to be used. Also, in a letter submitted in evidence, written about a week before the Corona fatality that led to OSHA's inspection, Superintendent Harding brought Gillie's attention to Rule 4 and added, "Switch crew members are not forced to work with B/O [bad order] equipment as stated in your letter." The judge found that the testimony of Warner, Thedford, and Gillie about yardmasters actively encouraging safety rule violations "was flatly contradicted by Inland's witnesses in many material respects; and when questioned on cross examination by Inland's counsel, [Warner's, Thedford's, and Gillie's] account of particular events created a strong impression that they were giving a wrong coloring to material facts so as to deprive them of credit." The judge also discredited Gillies testimony generally, as discussed above. Like the Secretary, we are troubled by an implication in the judge's decision that the employees' claims of harassment are not credible because their testimony showed only that they broke work rules to gain from supervisors longer lunch periods and earlier quitting time. Such actions by supervisors would tend to undermine the company's safety program. However, all the credibility findings concerning Warner, Thedford, and Gillie are adequately explained on the basis of matters peculiarly observable by the judge--the witnesses' demeanor on the stand and their manner of responding on cross-examination. The judge's findings also rest on the contrary testimony of Inland's witnesses, whom the judge also observed. We would have preferred that the judge had specifically identified the supervisors' testimony that conflicted with the testimony he discredited. However, we have reviewed the entire record and have noted above testimony by yardmasters Westerfield and Murray that contradicts the employees' testimony that supervisors encouraged violations. We cannot say that the judge erred in crediting the supervisors' testimony rather than that of the switchmen.[[15]] We therefore accept the judge's credibility determinations. _C. Kaufman_, _supra_; _cf_. _Asplundh Tree Expert Co_., 79 OSAHRC 109/A2, 7 BNA OSHC 2074, 2078-79, 1979 CCH OSHD ¶ 24,147, pp. 29,346-47 (No. 16162, 1979) (generalized criticism of a witness as not credible is insufficient). We further agree with the judge's findings that certain employee testimony of supervisory coercion is simply insubstantial. For example, the only specific incident that Connelly discussed in sufficient detail to allow us to evaluate what he considered "harassment" seriously undermines his allegation, as the judge found. That was the incident involving former yardmaster Lambert, discussed previously. The sum and substance of that incident is that Connelly concurred that the cars could be unloaded at the mill before being repaired, after Lambert "implored" him to do so because the cars were needed there. Connelly also testified that Lambert had had a heart attack "and was under very severe pressure to stay on the job, and we bent over backwards to keep Jim Lambert without stress or strain." We agree with the judge that this incident shows no harassment or coercion by Lambert. Switchman Dills also testified that he had violated safety rules to "[g]et the job done" and "[g]et in earlier, get better breaks" from the yardmasters. However, the one specific instance he recounted in this regard does not support the implication that yardmasters would reward him for breaking rules. Dills testified that "3 weeks ago I was working a little slower, I didn't feel that good that evening, and I was taking my time on my job, the Yard Master got ticked off because I wasn't working any faster than I was. And he put me on a different job" which Dills considered to be a punishment because "[i]t's a little extra work . . . . there's not that much time, where you can get a good break on that other job." However, when Dills told the trainmaster that he was not feeling well he was permitted to go to the clinic. He took two days off work and returned without further incident. The judge correctly found that Dills's testimony that he was "harassed" in that instance had "no substance." Lastly, the judge discredited the testimony of Dills that when he had been an hourly foreman in the Transportation Department for about one and one-half years ending in early 1978, other supervisors falsified safety contract cards by writing down safety contacts with employees that never had occurred. After testifying to that effect, Dills admitted it was possible that those yardmasters had made the safety contacts before he saw them fill out the cards, and that the only falsifications he could actually speak to were his own. We uphold the judge's finding that Dills' testimony regarding the possible falsifications by other yardmasters "lack[ed] credibility." Insofar as Dill's falsifications are concerned, he admitted that Inland had taught him the correct way to fill out the cards, and that he could not confirm that any yardmasters knew he falsified them. Indeed, Dills testified that he would not want Harding to find out because "[i]t's breaking a rule, his rule." Dills' admissions constitute the only proof of violations by any foreman or supervisor in this case. They were not safety rule violations, but concern a tangential matter--safety contacts with switchmen in the yards. This idiosyncratic misconduct of a single hourly foreman over a year before the citation is too frail a need to support this citation. _See_ _Pennsylvania Power & Light Co_., 737 F.2d 350 (3d Cir. 1984). Finally, the compliance officer testified that during his inspection, yardmaster Murray went between two cars and manually operated a pin lifter in his presence. However, Murray testified that he had done so because he understood that the compliance officer had asked him "to demonstrate what we felt caused [Corona's] accident." Murray testified that the compliance officer then "asked me if it could be done without being between the draw bars" and Murray demonstrated that also. He testified that these demonstrations had occurred in the presence of a Mr. Jones of the safety department and Assistant Superintendent DuVall as well. Only later did the compliance officer discover that lifting pins manually was a safety violation. Murray's testimony adequately explains the reason for demonstrating the manual lifting of a pin. He did it to satisfy what he understood to be a request by an OSHA inspector during the inspection. Exposure to hazards due to complying with an OSHA inspector's perceived request is not grounds for issuance of a citation. _Cf_. _Brown-McKee, Inc_., 80 OSAHRC 34/A2, 8 BNA OSHC 1247, 1249, 1980 CCH OSHD ¶ 24,409, p. 29,736 (No. 76-982, 1980) (violation cannot be based on exposure of employer's walkaround representative during inspection); _Bechtel Power_ _Co_., 79 OSAHRC 34/A2, 7 BNA OSHC 1361, 1365 n.7, 1979 CCH OSHD ¶ 23,575 at p. 28,576 n.7 (No. 13832, 1979) (same). We accept the judge's determination that the evidence did not preponderate in the Secretary's favor on the issue of the adequacy of Inland's efforts to enforce its safety program.[[16]] Thus, the Secretary failed to prove a section 5(a)(1) violation with regard to the aspects of Inland's safety program at issue--its rules for reporting defective railcars and uncoupling railcars. Because the Secretary failed to prove a section 5(a)(1) violation as to either citation item, we vacate the citation. The judge's decision is affirmed. FOR THE COMMISSION Ray H. Darling, Jr. Executive Secretary DATED: JUL 30, 1986 ------------------------------------------------------------------------ The Administrative Law Judge decision in this matter is unavailable in this format. To obtain a copy of this document, please request one from our Public Information Office by e-mail ( lwhitsett@oshrc.gov ), telephone (202-606-5398), fax (202-606-5050), TTY (202-606-5386). FOOTNOTES: [[*]] Commissioner Wall did not participate in this decision. [[1]] Section 5(a)(1) states: Each employer-- (1) shall furnish to each of his employees employment and a place of employment which are free from recognized hazards that are causing or are likely to cause death or serious physical harm to his employees . . . . [[2]] The Authorized Employee Representative, which also seeks enforcement of the citation, joined in many of the Secretary's arguments. Where both of these parties advance an argument, we will identify it as the Secretary's argument, for convenience. [[3]] The Secretary refers to the hazard at issue as both "inadvertent" and "uncontrolled" movement, and former Commission Cleary's direction for review used the term "uncontrolled" rather than "inadvertent." We conclude that in the circumstances here there is no substantive difference between the two terms. The Secretary also argues that the issue of whether a recognized hazard existed is not before us. The judge found that there was a recognized hazard of "inadvertent movement of railroad cars," and the correctness of that finding was not mentioned in former Commissioner Cleary's direction for review. However, the Commission may still rule on the issue after the parties are afforded any necessary opportunity to submit briefs on it. _See_ _Hamilton Die Cast, Inc_., 86 OSAHRC __ /__ , 12 BNA OSHC 1797, 1986 CCH OSHD ¶ (No. 83-308, 1986) (Commission may consider issues outside the four corners of a direction for review for the sake of fairness and uniformity of adjudication, after parties have been afforded any necessary opportunity to brief issues). All the parties specifically addressed the issue on the merits in their briefs. Moreover, the existence of a recognized hazard, properly defined, is an essential element of a finding of any section 5(a)(1) violation. In these circumstances, we will address the recognized hazard issue. [[4]] _In Wheeling-Pittsburgh Steel Corp_., 81 OSAHRC 102/A2, 10 BNA OSHC 1242, 1981 CCH OSHD ¶ 25,801 (No. 76-4807 1981), _aff'd without published opinion_ 688 F.2d 828 (3d Cir. 1982), _cert. denied_ 459 U.S. 1203 (1983), the Commission found that certain unidentified "other steel companies" use handbrakes on their railcars. 10 BNA OSHC at 1246 & n.4, 1981 CCH OSHD at p. 32,244 & n.4. However, that decision did not indicate how many steel companies use handbrakes. In this case there was testimony that at least seven steel company plants had received pugh ladles with handbrakes from one manufacturer. However, the record does not explain why these plants received pugh ladle cars with brakes, nor does it show that any of the plants actually used the brakes. Indeed, Inland's expert witness, Martin, stated that one of the steel companies that receives cars with brakes, Bethlehem Steel, operates in the same manner as does Inland, that is, it does not use brakes. Other steel plants in addition to Inland had pugh ladle cars that were not equipped with brakes. Thus, the record here does not establish that the steel industry generally recognizes that blocks alone will not prevent injury to employees. [[5]] The Union argues that Inland's rules should be more specific as to when railcars should be secured. It points to rules of various interstate common carriers as models. However, we find that the common carriers' rules are not models that Inland can be expected to follow. In any event, those rules were not actually shown to give more specific guidance than the training Inland gave employees in blocking. Most of the safety rules submitted in evidence by the Secretary contain general statements like that of the Elgin, Joliet & Eastern Railway Co.: "When cars are left on any track, sufficient hand brakes must be set to prevent cars moving . . . . " Inland's Rule 43 gives comparably specific guidance since it requires that railcars be blocked before employees step between them "if there is any question of the cars not being at a complete standstill." Also, Inland trains its switchmen in the different types of blocking and when to use them, and instructs employees to block cars so that there will be no movement after they are set. [[6]] The specific safety problems of setting and releasing handbrakes were explained by Martin. Typically, applying or releasing handbrakes requires the switchman to climb partway up a metal ladder on the side of the car, then step to a similar adjacent ladder on the end of the car and climb up that ladder to a narrow metal platform (perhaps 12" wide and 20"-30" long) to operate the handbrake. The switchman then sets or releases the handbrake, typically by turning a brake wheel or rotating a pump handle and retraces his steps down the ladder to detrain. Martin noted that the hazards of slippery stirrups and other surfaces would be increased in adverse weather conditions, or where there was grease on the stirrups. Martin also noted that brakes can malfunction due to chains breaking or other factors and that a railcar can move inadvertently--even run away from a train to which it was attached--while handbrakes are being applied or released, creating dangers to an employee attempting to leave the train by a ladder from the handbrake platform. Photographs appearing in the safety manual of the Elgin, Joliet and Eastern Railway Company, an interstate carrier, clearly support the judge's conclusion that switchmen face a "precarious situation" when operating handbrakes. [[7]] OSHA's compliance officer, Gigli, also gave opinion testimony tending to show that handbrakes should be required. However, Gigli had no personal experience with rolling railroad stock or steel railroads. Thus, his opinions are entitled to no greater weight that those of Sage and Barnes. The Secretary's evidence on the need for handbrakes did not preponderate over Inland's contrary evidence. [[8]] In his brief he notes that if the Commission finds that "such a procedure does create a hazard," the Secretary "only seeks to require the use of brakes to secure stationary railcars." The FRA injury statistics previously referred to show 246 injuries to employees while operating handbrakes on moving cars in yard service. A hazard therefore is clearly presented. [[9]] An explanation of certain terms used in the citation is helpful. The uncoupler devices, commonly called "pin lifters," permit railcars to be uncoupled without an employee stepping between the cars and facing the risk of railcar movement. The pin lifter consists of a lever extending from beneath the coupling mechanism ("knuckle") at the end of the railcar to the outer edge of the car. To uncouple two railcars, the employee pushes down on the lever, causing the other end of the lever to rise, pushing up the "pin" in the knuckler and unlocking the knuckle. The "stirrups" are the rungs of the metal ladders on the sides and ends of the railcars, and the "handholds" or "grab irons" are metal handles that employees grasp while climbing on and off the railcars or riding on them. The "drawbar" is the metal shaft, one end of which forms the knuckle and the other end of which attaches to the railcar's frame. The "yoke" attaches to the end of the railcar and acts as a collar around the "drawbar," holding it in place. [[10]] The Secretary and the Union rely on Commission cases involving safety standards promulgated under section 5(a)(2) of the Act, placing a burden on the a _prima_ _facie_ case. Those cases are inapposite under section 5(a)(1). The Secretary retains the burden of establishing by a preponderance of the evidence the existence of a feasible and useful means of abating recognized hazards in a section 5(a)(1) case. _See_ _Donovan v. Royal Logging_, 645 F.2d 822, 828 (9th Cir. 1981). [[11]] Yardmasters are the switchmen's supervisors and are responsible for the proper operation of their respective yards. The trainmaster on each shift is in overall charge of rail operations throughout Inland's facility. "Rip tracks" are repair tracks in the yard where bad order cars are to be directed and taken out of service until repaired. "Foreign cars" are railcars from outside railroads. "I.H.B." apparently refers to the Indiana Harbor Belt Railway and the "J" apparently refers to the Elgin, Joliet and Eastern Railway Company, which are outside railroads. [[12]] The Union notes that there were no written instructions informing employees what constitutes a defective car and argues that shows inadequate communication of Inland's safety rules. We disagree because that argument over-emphasizes formal aspects of Inland's safety program at the expense of its substance. The lack of written instructions was not shown to decrease employee awareness of the defects to be reported or reduce employee compliance with reporting procedures. _See_ _Jones & Laughlin Steel_ _Corp_., 82 OSAHRC 34/A2, 10 BNA OSHC 1778, 1982 CCH OSHD ¶ 26,128 (No. 76-2636, 1982). [[13]] We reach the same conclusion with respect to Connelly's testimony, which the judge did not address, of another instance when a great deal of scrap loading required the use of an unusual number of cars, including bad order cars. [[14]] The fact that Inland's repair tracks were full tends to establish that Inland's rule requiring the reporting and removal from service of defective railcars was adequately implemented. Conversely, however, it is equally obvious that the ultimate effectiveness of the rule depends as well on the adequacy of Inland's car repair procedures. Nevertheless, the Secretary's citation does not directly allege any inadequacy in Inland's car repair program itself. The record shows that in response to Gillie's request in October 1978 for an improved repair program, Inland assigned additional personnel and a second supervisor to the repair shop and established two additional repair tracks. At the time of the inspection, Inland each month was repairing between 750 and 800 cars, almost one-third of its entire fleet. In a memorandum written to DuVall in June 1979, after the inspection, Thedford, then union safety steward, expressed approval of car repair efforts during one week in that month when 181 cars were repaired. In referring to this amount of car repair, which is equivalent to a monthly repair rate of 785 cars, Thedford stated, "I think our [s]upervisors are doing a fine job in helping us have safe equipment to work with." Although Thedford also concluded that a large number of bad order cars remained, his memorandum recognized rule 4 and stated that employees can assist in reducing the incidence of bad order cars by reporting defects as required by that rule. Accordingly, Thedford apparently considered Inland's repair facilities to be capable of handling the number of defective cars that in his view should be reported under Rule 4. [[15]] Former switchman Connelly also testified generally that yardmasters who he did not identify threatened to shorten his lunch period or give him the least desirable jobs around the yard if he did not keep defective cars in service as needed in scrap operations. Like the testimony of Warner, Thedford, and Gillie Connelly's statements are contradicted by the testimony of Inland's supervisors. [[16]] The previous accidents involving Inland switchmen do not prove inadequacies in the elements of Inland's safety program at issue here. The Secretary relies on the Corona fatality as well as a fatality and several injuries that occurred in the 1960's to establish inadequacies. However, Corona, who was crushed between the drawbars of two railcars, had been instructed in Inland's safety rules, had correctly answered the questions on Rule 43 in his switchman's test, and had been orally reprimanded by trainmaster Webber three days before the fatality for violating Rule 43, as noted above. Yardmaster Westerfield testified that Corona told him the morning he died, "Don't worry, I know Rule 43." There is no evidence that Corona was not properly instructed in Inland's safety rules or that Inland condoned his violation of those rules. It also should be noted that switchman Thedford, who blamed yardmasters for failure to make sure that bad order cars were removed from service in certain cases was on duty in that one of the cars involved had a missing pin lifter, which may have led Corona to go between them, where he was crushed. As to the fatality and injuries that occurred in the 1960's, these also fail to show inadequacies in Inland's safety program at the time of the inspection or citations. The Secretary notes that a broken toe was suffered by switchman Bastardo in 1969 when a stirrup broke on a car he was riding and his foot fell under a wheel. However, the Secretary did not show that Inland should have detected a defect in that stirrup in advance. That incident establishes nothing about the adequacy of Inland's safety rules. A 1967 injury to switchman Schutz occurred while he was standing between cars inspecting a defective coupling mechanism. However, that injury predated Rule 60, which prohibited going between railcars having no pin lifters. A 1964 arm injury to switchman Holland, the result of being pinned between railcars while attempting to open a coupler, was due to his violation of Rule 43, according to Inland's accident report. A 1964 hand injury to coal bridge helper Wallace while setting blocks on a coal tripper car, involved specialized coal handling equipment and is unrelated to Inland's railroad operations, as the judge found. The cause of a fatality in 1960, in which an employee, Stubbs, was pinned between railcars, was not established. Thus, these accidents do not prove that Inland had an inadequate safety program.