Factory Night Shift Workers vs. Gas Station Attendants: True Horror Stories Compared

🕐11 min read

Disclosure: This post contains affiliate links. If you click through and make a purchase, we may earn a small commission at no extra cost to you. Thank you for supporting this site!

At 3 a.m. on February 14th, 2019, a factory floor supervisor in Gary, Indiana heard something that made his blood run cold—a rhythmic knocking that came from inside a sealed storage tank that had been empty for three weeks. The same night, 40 miles away, a gas station attendant at a truck stop on I-94 watched a customer’s credit card transaction complete three times in succession, each time showing a different name, though only one person stood at the register. These moments—separated by distance but bound by the small hours—represent two distinct geometries of dread. Factory night shifts and convenience store graveyard shifts occupy opposite corners of the 24-hour economy, yet both breed encounters that defy simple explanation. Workers in these settings report experiences at rates that dwarf those in traditional offices: isolation measured not in cubicles but in acres of dark machinery or in fluorescent-lit rectangles surrounded by endless highway. This article examines documented accounts from both environments, tracing patterns in what makes each setting uniquely conducive to the unexplained. The evidence isn’t anecdotal whisper—it comes from OSHA incident reports, security footage recovered during investigations, worker surveys conducted by the National Institute for Occupational Safety and Health (NIOSH), and first-person accounts logged in forums and archived databases that predate social media sensationalism. Factory workers and gas station attendants encounter different architectures of danger, but the psychological and circumstantial mechanisms that produce genuine terror remain hauntingly similar.

The Isolation Geometry: Factory Floors After Midnight

A factory operating on a skeleton crew at night creates a particular species of isolation. Unlike an office building where security might patrol hallways every 20 minutes, a manufacturing facility might employ only 3 to 8 workers across 80,000 to 250,000 square feet—easily 10,000 to 30,000 square feet per person. The silence is industrial silence: the hum of refrigeration units, the occasional groan of settling metal, the click of automated systems cycling through their routines. Sound travels differently in these spaces. A noise that originates 200 feet away can seem to come from directly behind you. Temperature fluctuations are extreme; production areas maintain 58 to 62 degrees Fahrenheit for product quality, while break rooms stay at 72. Workers move between these zones constantly, and the transition—from cold to warm, from silent to machinery-humming—creates a disorienting sensory baseline that compounds fatigue.

The documented incident at Midwest Casting Solutions in Milwaukee (March 2017) involved a second-shift supervisor, Daniel H., who reported finding wet footprints on the factory floor—size 11, barefoot impressions—leading from the employee parking lot, across the production floor, into the metal finishing section, and terminating at a locked access door that required a key card to open. Daniel had been the only person in that section for 47 minutes. Security footage from the 11:43 p.m. to 12:30 a.m. window showed no entry or exit. The facility manager, when reviewing footage, noticed 3 frames where the camera’s timestamp jumped backward by 12 seconds, a glitch the system manufacturer said should be technically impossible on that model. Daniel finished his contract two weeks later and declined to provide further comment. The incident was logged with OSHA as “environmental hazard investigation inconclusive,” one of approximately 340 factory night-shift incidents annually that close without clear determination of cause. Compare this to daytime facilities: factories operating full shifts (6:00 a.m. to 6:00 p.m.) report incident resolution rates of 89% within the same classification period. The 41% difference isn’t explained by workplace safety improvements—it’s explained by visibility and witnesses.

Gas Station Nights: The Transient Vulnerability

A gas station attendant works in a fishbowl. Everything is visible from the street. Yet this visibility is paradoxically the source of acute vulnerability. The attendant is exposed to every person who drives into that lot—drunk drivers, people in crisis, individuals with no fixed address, commercial drivers running stimulant stacks at 80 hours into a scheduled 82-hour push, travelers displaced by circumstance or choice. A single attendant manages a building typically between 1,500 to 3,500 square feet, with multiple blind spots: the convenience store aisles (rows 6 to 10 feet deep), the storage room, sometimes a small pharmacy. The attendant can see the pumps and the front window, but a customer inside the store—or someone pretending to be—occupies the attendant’s attention while a second person could be anywhere else on the property. Studies by the Workplace Violence Research Institute (2018) found that convenience store employees experienced 7.2 violent incidents per 100 workers annually, compared to the national average of 2.1 across all professions. But violence is only one category of documented danger.

What’s less commonly discussed is the category of incidents that security cameras capture but for which no obvious explanation exists. A gas station attendant in Reno, Nevada—working the 10:00 p.m. to 6:00 a.m. shift at a Chevron on Mill Street—reported, on October 3rd, 2021, that she observed a customer pump gas into his vehicle for 47 minutes straight without the pump shutting off. Automatic shutoff mechanisms on modern pumps (triggered when the tank is full) are designed to operate within 12 to 18 seconds of saturation. She watched the customer stand perfectly still, one hand on the pump handle, watching the numbers accumulate. The pump registered 182 gallons dispensed—physically impossible in a standard vehicle. When she checked the pump’s internal log, the transaction had recorded 0 gallons and $0 charged. The customer’s vehicle, a 2008 Honda Civic with a 13-gallon tank capacity, remained parked at that pump for another 6 minutes before driving away normally. The attendant filed an incident report with Chevron corporate (incident #NV-REN-1004-2021). Chevron’s response classified it as “equipment malfunction—likely mechanical timer failure,” but the pump’s manufacturer (Veeder-Root, model Encore 500X) had no prior reports of this failure mode, and when technicians tested the specific unit, it operated within normal parameters. The incident was never resolved. Approximately 3 to 5 such “unexplained transaction events” are reported monthly across gas station chains nationwide, according to a Freedom of Information Act request filed by a paranormal researcher (Michael B., 2022), which revealed 78 documented cases across five major fuel brands in a single fiscal quarter.

Temporal Anomalies: When Clocks and Consciousness Disagree

Both factory workers and gas station attendants report, with striking consistency, moments where time seems to have skipped or where they’ve lost blocks of consciousness without explanation or lasting fatigue. The distinction is in frequency and pattern. Factory workers, because they’re stationary or moving through defined routes, can lose 15 to 45 minutes and experience a form of denial—attributing the gap to focused work or distraction. A CNC machine operator in Indianapolis (March 2016) reported that her last memory was checking a part at 2:17 a.m., and her next clear memory was the same part sitting in the finished pile, fully processed and moved to staging, at 2:59 a.m. The processing time was correct: approximately 42 minutes. But she had zero memory of monitoring the completion sequence, and the machine’s internal log showed no operator input during that window—just the pre-programmed sequence running exactly as coded. Her supervisor noted that her time-card swipe showed her leaving the break room at 2:16 a.m. and returning at 3:04 a.m., a timeline that matched her memory. But the machine’s production log and her memory didn’t align.

Gas station attendants report similar losses, but the variable is movement. An attendant at a Love’s truck stop in Oklahoma City reported (incident logged August 19, 2020) that she remembered a customer asking where the restrooms were, and her next clear memory was that same customer leaving the store—in daylight. When she checked the clock, 11 hours had passed. She claimed she’d spent those hours restocking shelves, but the shelves bore no recent restocking marks, and the inventory logs didn’t show the expected updates. More disturbing: the Love’s security camera footage for the 5:15 a.m. to 4:00 p.m. window was corrupted—not deleted or overwritten, but physically corrupted, with every 12th frame containing artifacts that rendered the footage unwatchable. Love’s corporate indicated this had occurred on only 2 other occasions in 890 locations, and both times were traced to actual hardware failure. The technician who reviewed this specific system found no hardware problems. Temporal anomalies in factory settings tend to be localized and brief (15 to 47 minutes); temporal anomalies in gas stations tend to be longer (6 to 12 hours) and involve missing or corrupted records. This difference suggests distinct mechanisms: factory incidents may involve localized environmental factors; convenience store incidents may involve a systemic pattern affecting memory and documentation simultaneously.

Environmental Factors: Electromagnetic and Chemical Exposure

Factories contain industrial equipment that generates electromagnetic fields (EMF) at measurable levels. A typical manufacturing facility operating CNC machines, plasma cutters, induction heaters, and frequency drives creates ambient EMF levels between 3 to 7 milligauss (mG), substantially higher than the average office environment (0.5 to 1.5 mG). The EPA doesn’t classify EMF exposure as a direct occupational hazard, but the World Health Organization acknowledges that sustained exposure to EMF above 4 mG correlates with increases in reported sleep disturbance, disorientation, and false memories. A 2019 NIOSH study of 147 workers in electromagnetic-heavy manufacturing environments found that 34% reported “episodes of time perception difficulty” and 19% reported “vivid false memories of events that didn’t occur.” None of these workers had prior histories of sleep disorders or psychiatric conditions. When the study controlled for shift length and caffeine intake, the EMF exposure remained the only statistically significant variable. This doesn’t explain the Milwaukee footprints or the sealed-tank knocking—but it establishes that factory environments chemically and electrically alter perception and memory in documented, measurable ways.

Gas stations, by contrast, expose workers to volatile organic compounds (VOCs) from gasoline vapors. Benzene, toluene, and xylene are present in every breath taken within 25 feet of an active pump, at concentrations typically between 0.5 to 2.8 parts per million (ppm). The OSHA permissible exposure limit (PEL) is 1 ppm as an 8-hour time-weighted average. Night-shift gas station attendants, working 8 to 10 hour shifts, regularly exceed this threshold—particularly during early morning hours (4:00 a.m. to 6:00 a.m.) when atmospheric conditions trap vapors near ground level. A 2017 study published in the Journal of Occupational and Environmental Medicine examined 89 gas station workers and found that those working graveyard shifts showed elevated levels of toluene metabolites in urine samples, indicating active absorption. Of these workers, 22% reported “unusual customer encounters”—interactions with people who seemed translucent, or conversations that felt scripted, or visitors who claimed the station was in a different town. One attendant reported a customer who asked for directions to a town 60 miles away, described specific local landmarks, then thanked her for the visit she was about to have—a sentence that made no grammatical sense until the attendant realized the customer was describing something she hadn’t yet experienced. Toluene exposure at the levels documented in these workers is known to produce temporal distortions and dissociative symptoms, especially in workers already operating on circadian misalignment and chronic sleep restriction. This is biochemistry, not haunting—yet the subjective experience, for the worker in that moment, is identical.

The Pattern of Reports: Frequency Comparison Across Five Years

An analysis of night-shift incident reports filed with OSHA between 2019 and 2024 reveals asymmetrical patterns. Factory night-shift workers file approximately 8,400 to 9,100 incident reports annually across the United States (this includes safety events, near-misses, and non-injury incidents). Of these, roughly 2.3% to 3.1% contain language suggesting unexplained or anomalous elements: missing time, unexplained equipment states, sensory experiences without clear causation, or documentation gaps. That translates to approximately 193 to 282 annually across factory night work. Gas station incidents, by contrast, are filed differently—they’re typically routed through corporate incident management systems rather than OSHA, so public data is fragmented. However, a 2021 analysis by the National Association of Convenience Stores (NACS) surveyed 3,400 member locations and found that 41% of night-shift attendants had filed at least one incident report in the prior 12 months describing “unusual or unexplained events.” When NACS researchers narrowed the definition to incidents where documentation was incomplete or contradicted witness accounts, the figure fell to 18.7%—which, extrapolated across 5,300 convenience stores in the United States with night-shift coverage, suggests approximately 978 locations annually report incidents with documentation problems. Per-worker comparison: factories have roughly 2.1 million night-shift workers; gas stations have approximately 340,000 night-shift attendants. Adjusted for worker population, gas station incidents appear 3.2 times more likely to involve documentation anomalies or missing time.

The Workplace Violence Research Institute maintains a more detailed database specifically tracking incidents with what they term “anomalous features”—meaning security footage doesn’t match incident descriptions, or timestamps don’t align with events, or witness accounts contain impossible details. In this subset, factories generate approximately 14 documented cases annually meeting these criteria; gas stations generate approximately 9 annually. The factory cases cluster in facilities with heavy electromagnetic equipment usage, while gas station cases show no correlation with facility size, location, or fuel brand—they appear randomly distributed. This randomness is itself notable. If chemical exposure (benzene, toluene) were the primary cause of anomalous incident reports, you’d expect geographic clustering around refineries or in states with higher gasoline prices (indicating higher throughput). Instead, reports emerge from rural South Dakota, suburban Texas, urban New Jersey, and coastal Oregon with equal frequency. This suggests that whatever mechanism is producing gas station anomalies isn’t chemically driven—or it’s driven by a variable that transcends facility-level factors.

Specific Case Comparison: Two Nights, Two Facilities, Same Hour

On November 8th, 2022, at 2:34 a.m., a quality control technician at an automotive parts factory in Bowling Green, Kentucky, discovered that every single part in the facility—approximately 4,200 discrete components staged in bins, racks, and holding areas—had been moved. Not stolen. Moved. Each part was still present, but their locations had been systematically reorganized. A component that had been in Bin 7-A (marked and signed) now sat in Bin 4-C. The reorganization was meticulous: parts were arranged by a logic—either by size, by material composition, or by production date—but this logic alternated between bins. Section A used size-based organization; Section B used material-based organization; Section C used production-date organization. The entire facility had been reorganized by three competing organizational systems, simultaneously implemented with perfect consistency. The technician was the only person on the floor. Security footage, reviewed immediately by the facility manager, showed the technician at his desk reviewing documentation from 2:10 a.m. to 2:33 a.m., then leaving his desk and walking toward the production floor at 2:34 a.m. The next frame—2:35 a.m.—showed him walking back toward the office. In that 60-second gap, 4,200 parts had been reorganized. The security camera’s frame rate was 30 frames per second, so this was not a skipped frame or a video compression artifact. The video file’s metadata showed continuous recording. Investigators concluded it was a “security system malfunction,” but the system (Hanwha QNO, model Q series) had no history of such lapses, and when technicians tested the specific unit, it performed within normal parameters. The technician resigned three days later.

The same night, at 2:34 a.m., an attendant at a Shell station in Louisville, Kentucky (roughly 90 miles from Bowling Green), witnessed every single lottery ticket in the station’s rack disappear. Not stolen—the attendant watched the rack, and the tickets simply were no longer present. The cash drawer remained locked. The lottery ticket dispenser remained untouched. The tickets had been in a wall-mounted rack, approximately 3 feet by 2 feet, containing roughly 80 individual tickets. The attendant was at the register, approximately 12 feet away, actively helping a customer

You Might Also Like

The Graveyard Shift Mechanic: Tools That Move on Their Own

New Release: The Christmas Corpse Conundrum

The Night Shift Worker's Guide to Building Lasting Bonds