BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20251016T135157Z
LOCATION:Riverside East
DTSTART;TZID=America/Chicago:20251014T173000
DTEND;TZID=America/Chicago:20251014T183000
UID:HFESAM_ASPIRE 2025_sess131_POST144@linklings.com
SUMMARY:AI-Driven Heuristic Analysis: Enhancing Work Instructions with Gen
 erative Models
DESCRIPTION:Fernando Montalvo (Astrion); Promise Stephens (Eastern Univers
 ity, Aerodyne Industries); Phuoc Thai (University of Central Florida); and
  Zachary Bulger (Amentum)\n\nThe present study investigated the feasibilit
 y and reliability of using generative AI to conduct heuristic evaluations 
 of workplace instructions, comparing its performance to experienced human 
 evaluators. A custom GPT model, fine-tuned with examples and heuristic cri
 teria, evaluated nine sets of aerospace-based work instructions. The AI's 
 output included identifying weaknesses, suggesting improvements, scoring h
 euristics (1-10), and providing rationales for its input. Results showed m
 oderate-to-high agreement between the AI and human experts, with consisten
 t and reproducible AI scoring. Qualitative analysis confirmed the AI's abi
 lity to identify common weaknesses and offer relevant feedback, sometimes 
 even identifying issues missed by humans. While the AI provided adequate t
 ransparency, some explanations lacked detail, and minor discrepancies in j
 udgment necessitate continued human oversight. The research demonstrates t
 he potential of AI-driven heuristic evaluations to streamline assessment p
 rocesses and augment human analysis in high-risk industries, while acknowl
 edging the need for ongoing model refinement and improved transparency.\n\
 n
END:VEVENT
END:VCALENDAR
