Signatures of criticality in mining accidents and recurrent neural network forecasting model

Karan Doss, Alissa S. Hanshew, John Mauro

Research output: Contribution to journalArticle

Abstract

We report signatures of criticality in mining accident data obtained from the Mine Accident, Injury and Illness Report form (MSHA Form 7000-1). This work builds on the hypothesis that workplace accident statistics follow self-organized criticality (Mauro et al., 2018). “1/f noise,” a distinct feature of critical systems, is extracted from this database and is used to forecast accident trends using a long short-term memory (LSTM) recurrent neural network (RNN). The algorithm used for extracting this noise is applicable to data available in any standard worker's compensation database. We also report a Pareto distribution in the number of accidents in relation to employee mine experience, implying a strong correlation between experience and susceptibility to accidents.

Original languageEnglish (US)
Article number122656
JournalPhysica A: Statistical Mechanics and its Applications
Volume537
DOIs
StatePublished - Jan 1 2020

Fingerprint

Recurrent Neural Networks
Criticality
accidents
Accidents
forecasting
Forecasting
Mining
Signature
signatures
Model
1/f Noise
Pareto Distribution
Self-organized Criticality
Memory Term
Susceptibility
Forecast
statistics
Statistics
magnetic permeability
Distinct

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Condensed Matter Physics

Cite this

@article{2cc40eb76a2b40f1ac45069a0a45253b,
title = "Signatures of criticality in mining accidents and recurrent neural network forecasting model",
abstract = "We report signatures of criticality in mining accident data obtained from the Mine Accident, Injury and Illness Report form (MSHA Form 7000-1). This work builds on the hypothesis that workplace accident statistics follow self-organized criticality (Mauro et al., 2018). “1/f noise,” a distinct feature of critical systems, is extracted from this database and is used to forecast accident trends using a long short-term memory (LSTM) recurrent neural network (RNN). The algorithm used for extracting this noise is applicable to data available in any standard worker's compensation database. We also report a Pareto distribution in the number of accidents in relation to employee mine experience, implying a strong correlation between experience and susceptibility to accidents.",
author = "Karan Doss and Hanshew, {Alissa S.} and John Mauro",
year = "2020",
month = "1",
day = "1",
doi = "10.1016/j.physa.2019.122656",
language = "English (US)",
volume = "537",
journal = "Physica A: Statistical Mechanics and its Applications",
issn = "0378-4371",
publisher = "Elsevier",

}

Signatures of criticality in mining accidents and recurrent neural network forecasting model. / Doss, Karan; Hanshew, Alissa S.; Mauro, John.

In: Physica A: Statistical Mechanics and its Applications, Vol. 537, 122656, 01.01.2020.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Signatures of criticality in mining accidents and recurrent neural network forecasting model

AU - Doss, Karan

AU - Hanshew, Alissa S.

AU - Mauro, John

PY - 2020/1/1

Y1 - 2020/1/1

N2 - We report signatures of criticality in mining accident data obtained from the Mine Accident, Injury and Illness Report form (MSHA Form 7000-1). This work builds on the hypothesis that workplace accident statistics follow self-organized criticality (Mauro et al., 2018). “1/f noise,” a distinct feature of critical systems, is extracted from this database and is used to forecast accident trends using a long short-term memory (LSTM) recurrent neural network (RNN). The algorithm used for extracting this noise is applicable to data available in any standard worker's compensation database. We also report a Pareto distribution in the number of accidents in relation to employee mine experience, implying a strong correlation between experience and susceptibility to accidents.

AB - We report signatures of criticality in mining accident data obtained from the Mine Accident, Injury and Illness Report form (MSHA Form 7000-1). This work builds on the hypothesis that workplace accident statistics follow self-organized criticality (Mauro et al., 2018). “1/f noise,” a distinct feature of critical systems, is extracted from this database and is used to forecast accident trends using a long short-term memory (LSTM) recurrent neural network (RNN). The algorithm used for extracting this noise is applicable to data available in any standard worker's compensation database. We also report a Pareto distribution in the number of accidents in relation to employee mine experience, implying a strong correlation between experience and susceptibility to accidents.

UR - http://www.scopus.com/inward/record.url?scp=85072288183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072288183&partnerID=8YFLogxK

U2 - 10.1016/j.physa.2019.122656

DO - 10.1016/j.physa.2019.122656

M3 - Article

AN - SCOPUS:85072288183

VL - 537

JO - Physica A: Statistical Mechanics and its Applications

JF - Physica A: Statistical Mechanics and its Applications

SN - 0378-4371

M1 - 122656

ER -