Opinion
No. 22-60466
10-06-2023
DARLING INGREDIENTS, INCORPORATED, Petitioner, v. OCCUPATIONAL SAFETY AND HEALTH REVIEW COMMISSION; Martin Walsh, Secretary, U.S. Department of Labor, Respondents.
Juliana Connolly Gaige (argued), Jackson Lewis, P.C., Houston, TX, for Petitioner. John X. Cerveny, Occupational Safety & Health Review Commission, Washington, DC, for Respondent Occupational Safety and Health Review Commission. Jessica Matthis (argued), Heather Renee Phillips, U.S. Department of Labor, Office of the Solicitor, Washington, DC, Louise McGauley Betts, Esq., U.S. Department of Labor, Washington, DC, Seema Nanda, William W. Thompson, Office of the Solicitor, Occupational Safety & Health Division, Washington, DC, for Respondent Secretary, U.S. Department of Labor.
Appeal from the Occupational Safety & Health Review Commission, Agency No. XX-0240 Juliana Connolly Gaige (argued), Jackson Lewis, P.C., Houston, TX, for Petitioner. John X. Cerveny, Occupational Safety & Health Review Commission, Washington, DC, for Respondent Occupational Safety and Health Review Commission. Jessica Matthis (argued), Heather Renee Phillips, U.S. Department of Labor, Office of the Solicitor, Washington, DC, Louise McGauley Betts, Esq., U.S. Department of Labor, Washington, DC, Seema Nanda, William W. Thompson, Office of the Solicitor, Occupational Safety & Health Division, Washington, DC, for Respondent Secretary, U.S. Department of Labor. Before Jones, Clement, and Haynes, Circuit Judges. Edith Brown Clement, Circuit Judge:
Two people were killed while unclogging a machine at a chicken-rendering plant. When the Occupational Safety and Health Agency investigated, it found that the plant's "lockout/tagout" procedures did not "clearly and specifically outline" how to safely work on the machine, so it cited the plant's owner. Because we find no error in the decision to uphold those citations, we AFFIRM.
I.
Darling Ingredients, Inc. operates a chicken-rendering facility in Byram, Mississippi. One of the machines there is a "hydrolyzer," or a contraption that uses pressurized steam to break down "poultry parts," namely feathers and quills. The result is a digestible liquid known as "poultry meal" which Darling uses to make pet food.
Such a facility "cuts and grinds accepted animal carcasses into small pieces, to be blended[,] cooked," and "separated" into "[f]at, protein, and water," which is used to make products like soap, cooking oils, and animal feed. RENDERING MODULE, UNITED STATES DEPARTMENT OF AGRICULTURE, www.aphis.usda.gov/animal_health/carcass/docs/training/7-rendering.pdf.
Like any machine, the hydrolyzer requires upkeep. Specifically, chicken parts regularly build up in the hydrolyzer and lead to "blockage[s]." When that happens, pressure becomes trapped and "caus[es] the whole [machine] to stop." To get it running again, the hydrolyzer must be cleaned out. But to do that, the hydrolyzer's pent-up pressure must be released first. Apparently, that can be done in a number of ways. An operator can "shuttl[e] the gates," meaning he or she opens and closes a series of doors in the machine in a certain order, or unlatch a pressure-relief valve. A worker could also wait for the pressure to bleed off naturally, but that takes time.
On August 10, 2020, the Unit A hydrolyzer at Darling's facility became clogged, causing a pressure buildup. The machine's operator "shuttl[ed] the gates," but nothing happened. So he alerted his supervisor, and his supervisor called the maintenance team. Maintenance sent over three men—Fortenberry, Young, and Jackson. When they arrived on the scene, the trio tried the usual tricks—they shuttled the gates and opened the relief valve—but to no avail. So they cut the power to the hydrolyzer and "closed the steam valve" so that no more steam would flow into the machine. Then, with apparent hesitation, the team decided to remove a bolted, eight-inch flange from the hydrolyzer in order to release the trapped pressure.
But the removal of any part of the hydrolyzer may result in rapid depressurization. So, by removing the flange's bolts, the maintenance crew risked exposing themselves to the release of 2,000 pounds or more of pressurized air. That wasn't lost on the group—each time they took a bolt off, they would "jump back." Eventually, however, the "pressure overcame the threading" of the remaining bolts and the "flange burst open." When it did, "steam and hot material . . . spewed out of the machine," "covering" the three workers. Jackson and Fortenberry were gravely burned and died from their wounds.
When the Occupational Safety and Health Administration learned of Fortenberry and Jackson's deaths, they launched an investigation. The agency "took photographs, measurements, [and] conducted interview[s]." After the investigation, OSHA cited Darling for two violations of 29 C.F.R. § 1910.147, a "lockout/tagout" regulation. Lockout/tagout is an industry term for a safety system that tries to prevent the "unexpected energization or start up of [ ] machines" or the "release of stored energy" during "servic[e] and maintenance." 29 C.F.R. § 1910.147(b). Section 147 instructs employers to safeguard their employees from such dangers through isolation devices, protective equipment, regular training programs, and detailed written work procedures. See 29 C.F.R. § 1910.147.
Specifically, Darling was charged with failing to "clearly and specifically outline the . . . rules and technique to be utilized for the control of hazardous energy," § 1910.147(c)(4)(ii), and (2) "clearly and specifically outline the steps for shutting down, isolating, blocking and securing the machines or equipment to control hazardous energy" in its "energy control procedures," § 1910.147(c)(4)(ii)(B). In citing Darling, OSHA concluded that the maintenance crew was exposed to a "hazardous energy source"—the release of "[h]igh pressure steam" from the hydrolyzer—because Darling did not have any "specific procedure or steps" for the "employee[s] to reference" while working on the machine.
Darling challenged the citations before the Occupational Safety and Health Review Commission. At a trial before an Administrative Law Judge, Darling insisted that the maintenance team went off script. The company maintained that the group should have called a supervisor—as they were trained to do—or let the pressure bleed off naturally after their initial efforts failed. Darling claims that this was the established practice at the Byram facility and was clearly stated in the hydrolyzer's machine-specific lockout/tagout procedure. OSHA, on the other hand, argued that the crew's conduct stemmed directly from a lack of clearly written guidance in the company's lockout/tagout procedures. Specifically, "Step 6" of the procedure told the workers only that, to make the machine safe, they must "[r]elieve internal pressure." Per OSHA, Darling should have—as required by law—written a clearer, more instructive policy on locking out the hydrolyzer before working on it.
After the trial, the ALJ ruled in favor of OSHA, finding that (1) Darling did violate Section 147; (2) the violation was a repeat; (3) it was serious; and (4) Darling waived any "independent employee misconduct" defense. Darling appeals all of these decisions, save for the serious classification.
We note briefly that the ALJ concluded that "there is no question[ ] the violations were serious—death did occur and it was a probable consequence if an accident resulted from the violative condition." The former factor—the occurrence of a death—is not a consideration under the "seriousness" standard. Instead, in line with the court's latter conclusion, we ask whether there's a " 'substantial probability' that a particular violation could result in death or serious physical harm.' " Chao v. Occupational Safety & Health Rev. Comm'n, 401 F.3d 355, 367 (5th Cir. 2005) (citation omitted). But because Darling didn't appeal that issue, it's forfeited. See Rollins v. Home Depot USA, Inc., 8 F.4th 393, 397 (5th Cir. 2021).
II.
We are "bound by the administrative law judge's findings of fact" if they are "supported by 'substantial evidence.' " Austin Indus. Specialty Servs., L.P. v. Occupational Safety & Health Rev. Comm'n, 765 F.3d 434, 438 (5th Cir. 2014) (per curiam) (quoting 29 U.S.C. § 660(a)). Substantial evidence means "evidence as a reasonable mind might accept as adequate to support a conclusion." Id. It's a "deferential" standard. S. Hens, Inc. v. Occupational Safety & Health Rev. Comm'n, 930 F.3d 667, 682 (5th Cir. 2019). "[E]ven if the appellate court might have reached a different conclusion," it must "uphold [the] factual findings if a reasonable person could have found" similarly. Sanderson Farms, Inc. v. Perez, 811 F.3d 730, 734 (5th Cir. 2016) (citation and quotations omitted); Excel Modular Scaffold & Leasing Co. v. Occupational Safety & Health Rev. Comm'n, 943 F.3d 748, 754 (5th Cir. 2019) (ALJ decisions will be upheld "even if this court could justifiably reach a different result de novo") (citation and quotations omitted). As for "legal conclusions," they are reviewed for whether they are "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." S. Hens, 930 F.3d at 675 (citations omitted).
A.
We begin with Darling's merits-based challenge to the citations. To start, "[OSHA] has the burden of proving sufficient facts to support [a] citation." S. Hens., 930 F.3d at 675. That means they must show "by a preponderance of the evidence: (1) that the cited standard applies; (2) noncompliance with the cited standard; (3) access or exposure to the violative conditions; and (4) that the employer had actual or constructive knowledge of the conditions through the exercise of reasonable due diligence." Sanderson Farms, 811 F.3d at 735 (citing Sec'y of Lab. v. Jesse Remodeling, LLC, 22 BNA OSHC 1340 (O.S.H.R.C.A.L.J. June 2, 2006)).
After a trial, the ALJ found that OSHA established all four requirements to uphold the Section 147 violations. First, the regulation clearly applied to the conduct of the maintenance crew—"Darling admits that[.]" Second, Darling's employees were exposed to the risks of a lacking procedure, namely a "serious burn hazard," because of the violation. Darling does not contest that either. Third, Darling's procedure did not comply with Section 147's requirement that an employer "clearly and specifically outline" the methods for controlling hazardous energy. The ALJ rejected Darling's argument that its procedures must be "read in conjunction" with the hydrolyzer's manual and any training the team received. The procedure, the ALJ reasoned, must be sufficient in and of itself. But, as OSHA argued, the hydrolyzer's procedure was too broad and uninstructive. Finally, the ALJ found that Darling had actual or constructive knowledge that its lockout/tagout procedures violated Section 147. Specifically, OSHA proved that "Darling knew of the contents of its own [lockout/tagout] procedure, and also knew it had recently been cited for a similar [ ] violation." On appeal, Darling advances two arguments: (1) it complied with Section 147; and (2) it did not have knowledge of any violation. Considering the standard of review at play, we disagree.
1.
First, the ALJ found that Darling's lockout/tagout procedure did not satisfy the requirements of Section 147. Generally speaking, a lockout/tagout procedure must be at least detailed enough "to guide an employee through the lockout process." Gen. Motors Corp., 22 O.S.H. Cas. (BNA) 1019 at *8 (O.S.H.R.C. Dec. 4, 2007) (citation omitted). To comply with Section 147, a procedure must "inform the employee of the specific procedural steps to shut down and lock out a machine." Id.
In General Motors, the Commission found that a "brief[ ]" lockout/tagout procedure made up of generic commands (i.e., use "normal stopping procedure," ensure "dissipation of any stored energy") was not adequate. Id. Such language "lacks [ ] the specifics" required by Section 147. Id. In another case, Secretary of Labor v. Drexel Chemical Company, an employer's written procedures "f[e]ll far short" because they did not guide employees on "the type and magnitude of the energy, the method to control the energy, the shut down procedure, the energy isolating device and method, and the method to dissipate stored or residual energy." 17 O.S.H. Cas. (BNA) 1908 (O.S.H.R.C. Mar. 3, 1997) (emphasis added); see also Basic Grain Prod., Inc., 24 O.S.H. Cas. (BNA) 2024 at *9 (O.S.H.R.C.A.L.J. Nov. 5, 2013) (under Section 147, a lockout/tagout procedure must provide "methods to dissipate or restrain energy").
With that in mind, we turn to Darling's lockout/tagout procedure. The machine-specific procedure for the hydrolyzer is a ten-step process that is mandatory when "maintenance or servicing is done." After all, this process seeks to ensure that the machine is safe to work on. Consequently, the first few steps (i.e., Steps 1-5) are not surprising: a worker must notify anyone in the area of pending work and turn the machine off. Then, a worker must use "energy isolating devices," or "lockout[s]," to ensure that the machine isn't turned back on. The rub, however, is in Step 6. At Step 6, before working on the machine, a worker must "[m]ake all of the following sources of stored energy"—including "steam"—"safe by relieving pressure restraining, disconnecting, or discharging: Relieve internal pressure."
Darling has a general lockout/tagout policy that applies company-wide. That policy is not at issue. See 29 C.F.R. § 1910.147(c) (providing that, with some exceptions, machine-specific procedures must be created).
Image materials not available for display.
In reading Step 6, we're left with a simple question—what does "[r]elieve internal pressure" mean? According to Darling, once you reach Step 6, you are "to do nothing [ ] as that is the only way to relieve the internal pressure from the hydrolyzer." Consequently, Darling told its employees "all that they need[ed] to tell [them]." OSHA, on the other hand, argues that Step 6 is not detailed enough to instruct a worker on how to make the "sources of stored energy[y] . . . safe," much less how to "[r]elieve internal pressure." There is "no specific instruction . . . on how to do that," or anything "that even tells the operator . . . [to] go find a supervisor." We agree with OSHA.
Practically speaking, Darling's argument is flawed for a couple of reasons. First, there are steps that a worker can take besides waiting around. For example, they can shuttle the gates or open the relief valve. But those two well-established activities aren't written down in the protocol. Neither is the "next step" that you follow if those do not work, namely "contact [a] supervisor." In fact, Darling's personnel plainly admitted at trial that such commands are "not written in the procedure." In short, Darling's policy does not match up with its own current practices.
Second, doing nothing is doing something; if waiting was the right thing to do, there is no reason that Darling's procedure could not say that. In response, Darling contends that it should not be required to put "what ifs" and "troubleshooting procedures" in the lockout/tagout policy. But, by its own admission, sitting by is not a "what if"—it's the only thing to do. Yet it's not in the policy. That means Darling's current procedure is in direct conflict with the supposedly correct approach of sitting and waiting. After all, to "relieve pressure" is an active instruction.
But the shortcomings of Darling's policy go beyond practical considerations. Consider General Motors and Drexel Chemical. In both cases, the lockout/tagout policies were "overgeneralized document[s]" that, as a result, had "little to no practical utility for the end user." Basic Grain, 24 O.S.H. Cas. (BNA) 2024 at *10; Control of Hazardous Energy Sources (Lockout/Tagout), 54 Fed. Reg. 36644-01 (to be codified at 29 C.F.R. Pt. 1910). Ordinarily, "because the purpose of the lockout procedure is to guide an employee through the lockout process, [ ] general procedures are not acceptable." Drexel Chem., 17 O.S.H. Cas. (BNA) 1908 at *5. That is why procedures must be specific, "developed, documented and utilized." 29 C.F.R. § 1910.147(c)(4)(i); S. Hens, 2018 O.S.H. Dec. (CCH) ¶ 33662 at *5 (CMPAU Mar. 20, 2018).
Here, Step 6's command gives away the game: you're told to, but not how to, "[r]elieve internal pressure." That is fatal. At the very least, Darling's procedures should have told the crew how to "dissipate or restrain" energy stored in the hydrolyzer. Basic Grain, 24 O.S.H. Cas. (BNA) 2024 at *10. Instead, it simply states that the crew must "[r]elieve" pressure. That is not detailed enough. See S. Hens, 2018 O.S.H. Dec. (CCH) ¶ 33662 at *5 (CMPAU Mar. 20, 2018) (finding OSHA has "emphasized the importance of having [the] procedures in writing" and noting that "[n]one of the written procedures or training materials contain an explicit instruction to shut down and lock out the equipment after rinsing and foaming but before scrubbing," meaning the "document is not specific to any equipment and provides no further explanation. It is not a clear directive"). We agree with the ALJ—Darling's "procedures must say more," whether by providing a detailed method or even telling the team "to stop and wait . . . until the pressure dissipates . . . ." But Darling did neither. Consequently, the user is left with no clear guidance on how to safely control the release of hazardous pressure and steam from the hydrolyzer. Thus, there is substantial evidence to support the ALJ's finding of noncompliance, namely that Darling failed to "clearly and specifically outline the . . . rules and technique to be utilized for the control of hazardous energy." 29 C.F.R. § 1910.147(c)(4)(ii).
2.
Second, the ALJ found that Darling knew or should have known its policies didn't pass Section 147 muster. Under our caselaw, Darling must have "[known] of, or with [the] exercise of reasonable diligence could have known of the non-complying condition." Trinity Indus., Inc. v. Occupational Safety & Health Rev. Comm'n, 206 F.3d 539, 542 (5th Cir. 2000). The knowledge prong is focused on the "physical conditions constituting the violation," meaning "[t]he departure from OSHA standards, not the worker's injury, is the violation." S. Hens, 930 F.3d at 676, 679 (citing Calpine Corp. v. O.S.H.R.C., 774 F. App'x 879 (5th Cir. 2019) (per curiam)). And, under our caselaw, a "physical condition" can be an intangible thing or even the absence of something. See id. at 676. (discussing the lack of training as a violation).
Darling insists that it did not have knowledge of any Section 147 violation because it did not know the maintenance crew would remove the cover to the hydrolyzer's flange. But Darling misunderstands the focus of the knowledge prong. The company was cited for failing to "clearly and specifically address appropriate lockout, tagout procedures for steam trapped in [the] hydrolyzer during clog removal." It was not cited for the maintenance team's conduct. After all, the "physical conditions constituting the violation" was Darling's lacking written policy, nothing else. Id.
Concluding that Darling had knowledge of its insufficient written procedures is easy enough. The ALJ found that "Darling knew or . . . should have known of" them because "Darling knew of the contents of its own [ ] procedure, and also knew it had recently been cited for a similar [lockout/tagout] violation." There is substantial evidence in the record to support that finding.
Darling—as the master of its files—"created, reviewed, and [ ] revised" its lockout/tagout procedures. That includes creating and annually assessing, including in 2020, the hydrolyzer's machine-specific procedure. That alone may be enough to satisfy the knowledge requirement for a Section 147(c) violation. In Basic Grain, the Commission found that an employer had "direct knowledge" of its Section 147(c) violation because it had an employee who "was responsible for drafting and implementing the [lockout/tagout] policy." 24 O.S.H. Cas. (BNA) 2024 at *11. The Commission reasoned that the employer had "constructive knowledge" of the violation because its employee "could have performed a more diligent search of OSHA's website" and federal regulations "which would have placed her on notice that" the lockout/tagout policy "was not in compliance with the standard." Id.
Beyond the paperwork, though, Darling was aware that it might have problematic procedures company-wide. Just months before the citations at issue, OSHA cited Darling for the exact same violations of Section 147 at an Idaho plant. The year before, in 2019, Darling was forced to reassess all of its lockout/tagout procedures after a Tampa facility was similarly cited for insufficient written protocols. Ultimately, "employer knowledge [is] a fact-specific, practical inquiry" that "look[s] to company practice, the details of specific incidents, knowledge of supervisors imputable to the company, and commonsense inferences about what a company and its supervisors should know and do." S. Hens, 930 F.3d at 676. The factors for knowledge—including "company practice[s]" and "commonsense inferences"—clearly capture past violations. Id. Consequently, Darling's repeated violations for similar shortcomings likely should have alerted management—both company-wide and locally—to the need for changes to their lockout/tagout policies.
At the very least, the Byram facility—which has its own regional management—should have been tipped off by the fact that the hydrolyzer's written policies didn't comport with the actual practices of its employees. That's common sense. Id. Of course, if local supervisors knew (or should have known) of any shortcomings in the lockout protocols, that may mean Darling as a whole knew too. See Basic Grain, 24 O.S.H. Cas. (BNA) 2024 at *14 (imputing the "management team['s]" knowledge that employees were not trained to the employer as a whole). Consequently, for a host of reasons, we conclude that the ALJ's determination that Darling had knowledge of its Section 147 violations is supported by the law and substantial evidence.
B.
Now we briefly consider whether Darling's violations were repeat. Generally, the law discourages—via increased fines—employers from "repeatedly violat[ing]" OSHA regulations. 29 U.S.C. § 666(a). A violation is repetitive if, at the time it occurred, it was preceded by "a Commission final order against the same employer for a substantially similar violation." Bunge Corp. v. Sec'y of Lab., 638 F.2d 831, 837 (5th Cir. 1981). To start, OSHA must show that "the prior and present citations are for failure to comply with the same standard." Deep S. Crane & Rigging Co. v. Harris, 535 F. App'x 386, 390 (5th Cir. 2013) (per curiam); see also Bunge, 638 F.2d at 837. Then, the burden shifts to the employer to "rebut" that showing with "evidence of the dissimilarity of the conditions and hazards associated with the[ ] violations." Deep S. Crane, 535 F. App'x at 390. Then we ask "whether the prior and instant violations resulted in substantially similar hazards." Id. (citation omitted) (emphasis added).
Here, OSHA showed that Darling "was previously cited for violations of the same provisions," namely 29 C.F.R. §§ 1910.147(c)(4)(ii) and 1910.147(c)(4)(ii)(B), at a worksite in Idaho. In an attempt to show dissimilarity, Darling argues that because the citations involved "very different" machines—a hydrolyzer versus a "pneumatic air-powered" machine that "pushes cow carcasses"—the hazards at play were clearly different. We are not convinced.
Admittedly, the Idaho citations involved a different machine that almost certainly requires its own unique lockout/tagout procedures. That machine may even pose different hazards—say, a "crush injury" as opposed to the hydrolyzer's "burn" risk. But those considerations—which go to the nature of the machine in question—are not necessarily relevant here. Although meaningful, it is not even controlling that Darling was cited for the exact same violation. See Bunge, 638 F.2d at 837-38. Instead, we look to the hazards associated with Darling's violations. For example, in Deep South Crane this Court affirmed a repeat violation for inadequate training even though the two violations at issue "involved a different type of crane that caused a different injury." 535 F. App'x at 390. That is because "both violations were caused by the same hazard," namely that the employer "failed to adequately train a crane operator." Id.
The logic in Deep South Crane is instructive here. While the machines and injuries in question may have varied, the potential hazards that led to each violation were the same in each of Darling's cases. By failing to detail the lockout/tagout procedure in the Mississippi and Idaho facilities, Darling's employees were subjected to the same risk, namely that they would be inadvertently exposed to uncontrolled energy.
Darling's framing fails on first principles, too. The focus cannot be on the type of machine at issue. That would be simply too narrow of a reading of the law, and it would consequently render the "repeat" classification meaningless—a company would face increased penalties following prior violations only when it (1) violated the same exact law and (2) that violation was for the same exact machine.
Darling insists that the "record is replete with evidence of the lack of similarity between" the two citations and the "lack of specificity cited differ greatly between the two citations." Tellingly, however, Darling provides no record cites for those propositions. See FED. R. APP. P. 28(a)(8)(A). At the end of the day, the hazards are clearly similar, and Darling fails to disprove that similarity.
Darling has a high bar to overcome, too—"rebuttal may be difficult since the two violations almost have to be substantially similar in nature in order to constitute violations of the specific standard." Bunge, 638 F.2d at 837.
C.
Finally, we address whether Darling waived an "unforeseeable employee conduct" argument. We conclude that it did. Employers may assert an affirmative defense (although not recognized in any statute) that an OSHA violation was the result of "unpreventable employee misconduct." S. Hens, 930 F.3d at 678. Here, the ALJ found that "Darling failed to offer any evidence in support" of the defense at trial and did not "mention it in its post-trial brief, let alone point to any evidence in the record" to satisfy its burden. It is true that Darling did not push the defense at trial—other than a passing mention in its opening statement—or advance it in its post-trial memorandum. But, if a "party wishes to preserve an argument for appeal, the party 'must press and not merely intimate the argument during the proceedings before the [trial] court.' " Centerpoint Energy Hous. Elec. LLC v. Harris Cnty. Toll Rd. Auth., 246 F. App'x 286, 289 (5th Cir. 2007) (quoting Keelan v. Majesco Software, Inc., 407 F.3d 332, 340 (5th Cir. 2005)). Beyond that, Darling barely raises the argument before us—it presents a single page in its main brief and a few more in its reply brief. Still, it does not point the panel to any related or dispositive cases. See FED. R. APP. P. 28(a)(8)(A). Consequently, Darling failed to advance the argument before two different tribunals and has therefore waived the defense.
* * *
Considering the record and the deferential standard at play, we conclude that there is substantial evidence to support the ALJ's conclusions. Accordingly, we AFFIRM.