o > 
 
 ***** * 
 
 
 
 • * 
 
 :- -w •• 
 
 .' > 
 
 ♦«* 
 
 
 » " • • "*ri A . l ' » . ^ 
 
 
 » ^ 
 
 
 o. *^TV»' A 
 
 A 
 
 
 * **o« • 
 
 
 f •«: *W ;«& W .•»»: ^ • 
 
 « -.■< 
 
 
 
 • . » • A* *o ' A 
 
 /- '^o« •« 
 
 • ^ 
 
 ^«^ 
 
 iP-V. 
 
 
 /% 
 
 v "*b .*<> ..*»,-, <t 
 
 .0* V *^*KV A 
 
 , <p o «V^ ^> j* 
 
 
 "o^ 
 
 i0 V v'" '♦ *> 
 
 ^■^ 
 
 ■•-•• v^ 
 
 
 
 6°+ 
 
 V^ 7 ' ,*° 
 
 
 
 a0 v »LVL'* *> 
 
 k v-^ 
 
 • ^* y ^« °H 
 
 >v « 
 
 
 "0^ 
 
 
 ?^ 
 
 
 
 • ** % • 
 
 
 
 :• +*<t • 
 
 ^ u % --S 
 
 • ^o 
 
 
 
 
 

 ^°- 
 
 
 >»-... v^V v^V V-^> v*^ 
 
 
 
 
 ^ ''TV** A 
 
 .'... v^-y v-^v v^v v*> V* 
 
 
 
 ;^ 
 
 
 
 
 
 <$> " O H ' ^ 
 
 ^°- 
 
 s- '*. 
 
 
 
 
 Sr» -O^- »^&mil£xt\ .* ^ ^ ./»y^. r» . .sins' . . u • v J?^rn 
 
 
 
 ^\'- 
 
 
 ♦ <l^ r> * -vim- *• 
 
 « ^ 
 

Bureau of Mines Information Circular/1987 
 
 Roof and Rib Fall Accident and Cost 
 Statistics: An In- Depth Study 
 
 By Deno M. Pappas 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 
Information Circular 9151 
 
 ii 
 
 Roof and Rib Fall Accident and Cost 
 Statistics: An In- Depth Study 
 
 By Deno M. Pappas 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 Donald Paul Hodel, Secretary 
 
 BUREAU OF MINES 
 
 David S. Brown, Acting Director 
 
Library of Congress Cataloging in Publication Data: 
 
 
 Pappas, Deno M. 
 
 Roof and rib fall accident and cost statistics: an in-depth study. 
 
 (Bureau of Mines information circular ; 9151) 
 
 Bibliography: p. 19-20. 
 
 Supt. of Docs, no.: I 28.27: 9151. 
 
 1. Mine roof control. 2. Mine accidents. I. Title. II. Series: Information circular (United 
 States. Bureau of Mines) ; 9151. 
 
 TN295.U4 
 
 [TN317] 
 
 622 s [363. 1' 19622] 
 
 87-600117 
 
CONTENTS 
 
 Abstract 
 
 Introduction 
 
 Acknowledgments 
 
 Evaluation of roof-rib accident statistics 
 
 Review of accident fatalities, 1910-84 
 
 U. S. accident rates 
 
 State accident rates 
 
 Characteristics of U.S. roof-rib fall accidents 
 
 Seasonal patterns 
 
 Daily patterns 
 
 Mine attributes 
 
 Seam height 
 
 Mine size 
 
 Underground location 
 
 Mining method 
 
 The accident victim 
 
 Worker activity 
 
 Lost workdays 
 
 Type of injury 
 
 Cost analysis of roof -rib fall accidents 
 
 Total accident costs 
 
 Cost per accident 
 
 Comparison of accident costs based on degree of injury 
 
 Summary 
 
 References. 
 
 ILLUSTRATIONS 
 
 1. Underground fatality statistics, 1910-84 
 
 2. Roof-rib fall fatality statistics, 1931-84 
 
 3. U.S. roof-rib fall accident rates, 1980-84 
 
 4. U.S. roof-rib severity rates, 1980-84 
 
 5. Roof-rib accident rates, by State 
 
 6. Severity rates for roof -rib accidents, by State 
 
 7. Average number of roof-rib accidents with respect to month of occurrence. 
 
 8. Average number of roof-rib accident occurrences, with respect to time of 
 
 day 
 
 9. Roof -rib accident rates with respect to seam height of the mine 
 
 10. Roof-rib accident rates with respect to mine size based on the average 
 
 number of mine employees 
 
 1 1. Roof -rib accident locations 
 
 12. Frequency of roof-rib accidents with respect to type of mining method.... 
 
 13. Average number of lost workdays due to roof-rib accidents 
 
 14. Pie chart of roof-rib accident cost percentages, averages for 1980-84.... 
 
 15. Comparisons of costs per accident 
 
 Page 
 
 1 
 
 2 
 
 3 
 
 3 
 
 4 
 
 5 
 
 6 
 
 6 
 
 6 
 
 9 
 
 9 
 
 10 
 
 11 
 
 12 
 
 12 
 
 13 
 
 13 
 
 14 
 
 14 
 
 15 
 
 15 
 
 16 
 
 17 
 
 18 
 
 19 
 
 4 
 4 
 5 
 5 
 7 
 8 
 9 
 
 9 
 10 
 
 11 
 12 
 12 
 14 
 16 
 17 
 
11 
 
 
 TABLES 
 
 1. Roof -rib accident rates, longwall versus all other mining methods 
 
 2. Worker activity roof-rib accident rates, averages for 1980-84 
 
 3. Worker activity roof-rib fatality rates, averages for 1980-84 
 
 4. Distribution of roof -rib accidents with respect to days lost, averages for 
 
 1980-84 
 
 5. Comparison of roof-rib accident costs and all other types of accident costs 
 
 for 1 984 
 
 6. Comparison of costs per roof-rib accident based on degree of injury for 
 
 1983 
 
 Page 
 
 13 
 13 
 13 
 
 14 
 
 16 
 
 18 
 
 
 UNIT OF MEASURE 
 
 ABBREVIATIONS USED 
 
 IN 
 
 THIS REPORT 
 
 h 
 
 hour 
 
 MMst 
 
 
 million short tons 
 
 in 
 
 inch 
 
 st 
 
 
 short ton 
 
 min 
 
 minute 
 
 yr 
 
 
 year 
 
ROOF AND RIB FALL ACCIDENT AND COST STATISTICS: 
 
 AN IN-DEPTH STUDY 
 
 By Deno M. Pappas 1 
 
 ABSTRACT 
 
 The purpose of this Bureau of Mines study of U.S. roof and rib (roof- 
 rib) accident statistics and related accident costs is to define current 
 accident trends (1980-84) associated with fatal and nonfatal roof-rib 
 fall accidents. Data were retrieved from a data base containing all re- 
 corded U.S. mining accidents, then sorted and normalized utilizing a 
 computer software program. The statistics indicate that roof -rib acci- 
 dents have significantly declined in the 5-yr study period. Moreover, 
 they indicate that there have been increases and/or patterns of roof -rib 
 accidents associated with specific mine characteristics, such as seam 
 height, mine size, geographic location, and seasonal variations. Also, 
 roof-rib injury characteristics produced pattern changes involving 
 worker activity, lost workdays, types of injury, and severity of injury. 
 A conservative estimate indicates that there has been a 30% adjusted in- 
 crease in the cost of a roof -rib accident over the 5-yr study period. 
 
 'Research civil engineer, Pittsburgh Research Center, Bureau of Mines, Pittsburgh, 
 PA. 
 
INTRODUCTION 
 
 Underground coal mining has always been 
 recognized as a hazardous occupation. 
 Between 1931 and 1984, over 29,000 coal 
 miners lost their lives in underground 
 accidents. This represents an average of 
 one fatality for every 100,000 h worked 
 underground. Over 50% of these 29,000 
 fatalities were associated with roof-rib 
 fall accidents; this is a greater amount 
 than for any other class of accident. 
 Although there have been articles written 
 analyzing accident statistics, there has 
 been very little written specifically on 
 roof-rib fall accidents and costs. Such 
 is the intent of this paper. 
 
 Before looking at the statistics, the 
 term "roof-rib fall" needs to be defined. 
 Whenever an entry is developed in a coal 
 seam, the surrounding coal and rock mass 
 of the opening are no longer in equilib- 
 rium. The rock mass in the roof has lost 
 support from below, the floor rock no 
 longer has an applied load from above, 
 and the coal seam is no longer con- 
 strained along the sides (rib and face) 
 of the opening (1_). 2 If the roof is not 
 adequately supported, the surrounding 
 rock and coal may collapse or fall into 
 the entry and randomly strike underground 
 workers. In cases where an accident re- 
 sults from the fall of the face, it will 
 be considered a rib fall. Specifically, 
 the failure of the roof or rib may be at- 
 tributed to one or more of the following 
 factors: 
 
 1. Geologic anomalies. — Occur in the 
 roof or rib as faults, slips, joints, 
 rolls, clay veins, kettlebottoms , and 
 sand channels. 
 
 2. Effects of weathering. — Humidity 
 or temperature changes may cause the 
 
 ^Underlined numbers in parentheses re- 
 fer to items in the list of references at 
 the end of this report. 
 
 roof to deteriorate, which leads to roof 
 failure. 
 
 3. Stress conditions. — Stresses occur- 
 ring within the strata because of the ef- 
 fects of the overburden or the effects of 
 past geologic activity may result in a 
 roof-rib fall. For example, bumps or 
 bursts result in a sudden and violent 
 rupture of the supporting coal pillars 
 because the vertical unit loading of the 
 pillar exceeds the bearing strength of 
 the coal (2). High horizontal stress is 
 another stress condition that may cause a 
 roof-rib fall. The horizontal stress may 
 be in excess of the vertical stress, re- 
 sulting in open cracks along the entries 
 that may lead to failure of the roof. 
 
 4. Mining method. — The method employed 
 may initiate roof-rib failure such as in 
 longwall mining and retreat room-and- 
 pillar mining. 
 
 5. Scaling. — The occurrence of roof 
 failure caused by a worker barring down 
 the roof. Although, the failure of roof 
 was initiated by the worker, the accident 
 is still classified as a roof fall. 
 
 Most often roof-rib fall accidents 
 involve one or more of the preceding 
 factors, coupled with the fact that the 
 victim was under unsupported roof near 
 the face. The working face area is the 
 most hazardous area because the stresses 
 are being actively redistributed and 
 failure can occur instantly. Not in- 
 cluded as roof-rib falls are accidents 
 caused by haulage equipment knocking out 
 roof support because it is the motion of 
 the machinery that causes the accident. 
 Therefore, roof-rib falls can be related 
 to the unpredictable behavior of a rock 
 mass in transition from one state of 
 equilibrium to another and may be 
 initiated by several factors. 
 
ACKNOWLEDGMENTS 
 
 The author gratefully acknowledges Mrs. 
 Betty J. Hamilton, computer programmer 
 analyst, Theoretical Support Group, 
 
 Pittsburgh Research Center, for direction 
 in the use of the various computer soft- 
 ware packages. 
 
 EVALUATION OF ROOF-RIB ACCIDENT STATISTICS 
 
 To evaluate the roof-rib accident data, 
 it is first necessary to define the mean- 
 ing of the term "accident." For this 
 study, an accident is defined as any mis- 
 hap that results in a fatal or nonfatal 
 injury, including injuries that do not 
 result in lost workdays. It is important 
 to include nonfatal accidents data along 
 with the fatal accidents to obtain a 
 larger data base and a more complete pic- 
 ture of the extent of roof-rib fall acci- 
 dents. For every roof-rib fatality, 30 
 nonfatal accidents occur that result in 
 over 39,000 lost workdays (averaged) an- 
 nually. Therefore, to get a more com- 
 plete data base, the statistics include 
 both fatal and nonfatal roof-rib fall ac- 
 cidents for the 1980-84 period, except 
 where otherwise noted. 
 
 The following five types of accident 
 rates were calculated to evaluate roof- 
 rib accidents in U.S. underground coal 
 mines. These rates normalize the acci- 
 dents with respect to controlling factors 
 such as hours worked underground, produc- 
 tion, and size of work force, to give 
 a more concise measure of roof-rib 
 accidents. 
 
 1. Roof-rib fall accidents per 200,000 
 employee-hours worked underground. The 
 200, 000-h figure approximates the number 
 of hours worked by 100 full time miners 
 per year. 
 
 2. Roof-rib fall accidents per million 
 short tons of underground coal produced. 
 
 3. Roof-rib fall accidents per average 
 total number of underground workers. 
 This rate is actually a percentage of the 
 average underground work force injured in 
 these accidents. 
 
 4. Total number of lost workdays due 
 to roof-rib accidents per 200, 000 
 
 employee-hours worked underground. This 
 rate is known as the nonfatal severity 
 rate and indicates the seriousness of 
 nonfatal accidents. 
 
 5. Total number of lost workdays due 
 to roof-rib fatal, nonfatal, and perma- 
 nently disabling accidents per 200,000 
 employee-hours worked underground. This 
 rate is referred to as the overall 
 severity rate. It is similar to the non- 
 fatal severity rate except that it ac- 
 counts for accidents resulting in a fa- 
 tality or a permanent total disability. 
 Permanently disabling and fatal injuries 
 are each charged 6, 000 days (3)» 
 
 Raw data for this study were obtained 
 with the use of the Health and Safety 
 Analysis Center (HSAC) accident file 
 (from the Mine Information Systems of the 
 Denver Safety and Health Technology 
 Center of the Mine Safety and Health 
 Administration (MSHA)), which has on rec- 
 ord all reported U.S. mining accidents, 
 and was retrieved with the use of the Bu- 
 reau's program (HDBSEL). The records for 
 all roof-rib accidents that resulted in 
 fatal or nonfatal injuries (degree 1 to 
 6) were retrieved and transferred to the 
 RS/1 (Research Systems 1) software pack- 
 age designed by BBN Research Systems.-^ 
 The use of RS/1 permitted the retrieved 
 records to be tabulated, sorted, and 
 graphed. 
 
 Certain factors may be related to the 
 increase or decrease of accidents. How- 
 ever, an exact list of the factors in- 
 fluencing a pattern of roof-rib accidents 
 is nearly impossible to compile. Equally 
 
 •^Reference to specific products does 
 not imply endorsement by the Bureau of 
 Mines. 
 
difficult is the task of proving unequiv- 
 ocally that the factor is associated with 
 the accident statistics. There are many 
 hidden factors that may additionally af- 
 fect the compiled data such as lengthy 
 strikes, catastrophic disasters, and in- 
 consistencies in recording accident in- 
 formation. The accuracy of this study is 
 only as accurate as the accident records 
 entered into the data base. Consequent- 
 ly, this study emphasizes accident rate 
 trends that occurred over the 5-yr study 
 period rather than specific accident num- 
 bers or rates. Factors affecting acci- 
 dent trends are suggested, but are not 
 necessarily the only factors involved 
 and, in most instances, cannot be 
 definitely confirmed. 
 
 REVIEW OF ACCIDENT FATALITIES, 1910-84 
 
 A review of accident fatalities (fig. 
 1) over the past 74 yr reveals a dramatic 
 drop (approximately 96%) in the total un- 
 derground fatalities as well as in roof- 
 rib fall fatalities. Some of the major 
 reasons for these decreases are as 
 follows: 
 
 1. Decreases in accidents between 1920 
 and 1950 seem to coincide with drops in 
 production due to slowdowns in the econo- 
 my; e.g., the Great Depression of the 
 1930's, and the post-World War II reces- 
 sion of the late 1940's. 
 
 2. Mechanization of the mining indus- 
 try during the early 1950's, considerably 
 improved productivity and required a 
 smaller work force. As the number of 
 employee-hours decreased, so did the num- 
 ber of accidents (4^). 
 
 3. The increased use of roof bolts 
 starting in the 1950's. 
 
 4. Federal legislation, especially the 
 1952 Federal Coal Mine Act and the 1969 
 Coal Mine Health and Safety Act, which 
 promoted underground federal inspections 
 
 5. Bureau of Mines research efforts in 
 developments of the ATRS (automated tem- 
 porary roof support), FOPS [falling ob- 
 ject protective structures (canopies)], 
 methane drainage techniques, permissibil- 
 ity criteria for explosives, etc. 
 
 3,000 
 
 2,500 
 
 w 2,000 
 
 5 1,500 
 
 < 1,000 
 
 500 
 
 KEY 
 
 Total underground fatalities 
 
 Roof- rib fatalities 
 
 Roof-rib fatalities, % 
 
 
 M L-'iA 
 
 100 
 80 
 60 I 
 
 40§ 
 
 o 
 
 < 
 
 20 
 
 1900 1920 1940 I960 1980 2000 
 FIGURE 1.— Underground fatality statistics, 1910-84. 
 
 KEY 
 Roof-rib fatalities 
 
 Fatality rate : roof- rib accidents 
 
 per 200,000 employee- hours 
 • i i i I i i i 
 
 I930 I940 I950 I960 I970 I980 I990 
 FIGURE 2. — Roof-rib fall fatality statistics, 1931-84. 
 
 6. A greater awareness of underground 
 hazards promoted by mandated safety 
 training and supplemented by the union 
 and MSHA safety programs such as REAP 
 (roof evaluation accident prevention). 
 
 To obtain a more accurate picture of 
 the rate of roof-rib fall fatalitites, 
 the data were normalized, based on the 
 total employee-hours worked. From figure 
 2 it is quite clear that although the 
 number of fall fatalities started to 
 decrease in the 1940's, it was not until 
 the 1960's that the fatality rate (based 
 on employee-hours worked) actually 
 started to drop. Examination of roof-rib 
 accident percentages (fig. 1) shows that 
 a consistently high percentage (50%-70%) 
 of all the underground fatalities re- 
 sulted from roof-rib falls. It was only 
 recently that these percentages moderate- 
 ly dropped. During the early 1980's, 
 roof-rib fall fatalities averaged about 
 
40% of all underground fatalities, which 
 is still greater than any other type of 
 accident. 
 
 U.S. ACCIDENT RATES 
 
 Reviewing the various U.S. accident 
 rates (fig. 3) illustrates that roof-rib 
 accidents have consistently decreased 
 over the last 5 yr, hitting a new low 
 rate in 1983 and rebounding somewhat in 
 1984. These accident rates all confirm 
 that a decrease in roof-rib accidents oc- 
 curred: Accidents per million short tons 
 mined dropped 42%, accidents per 200,000 
 employee-hours worked decreased 17%, and 
 accidents per average number of workers 
 dropped 15%. 
 
 A probable reason for the decrease may 
 be due to the fact that between 1980 and 
 1984 there was a 28% decrease in the 
 average number of employees working un- 
 derground (_5). Even though the rates are 
 normalized linearly, the effects of a 
 smaller work force may have a dispropor- 
 tional effect in decreasing the number of 
 roof-rib accidents. Possibly, the mining 
 companies elected to keep their more ex- 
 perienced employees, which may have 
 resulted in a higher concentration of 
 safety orientated miners. 
 
 It is interesting to note that although 
 the number of workers decreased 28%, un- 
 derground coal production (short tons 
 
 KEY 
 • Based on production 
 a Based on employee-hours worked 
 ■ Based on total number of workers 
 
 < 
 
 a> 
 
 ° o 
 a. _c 
 
 1.4 
 
 1.0 
 
 0.012 
 
 .008 
 
 .004 
 
 
 
 1 
 
 A - 
 
 
 
 • 
 
 
 
 - 
 
 
 . 
 
 
 
 B - 
 
 - 
 
 
 " 
 
 
 
 ■ 
 
 
 
 
 
 C - 
 
 - 
 
 1__ 
 
 1 ' 
 
 
 1980 
 
 1981 
 
 1982 
 
 1983 
 
 1984 
 
 FIGURE 3.— U.S. roof-rib fall accident rates, 1980-84. A, ac- 
 cidents per million short tons of underground coal mined; B, 
 accidents per 200,000 employee-hours worked; C, accidents 
 per average total number of underground workers. 
 
 produced per year) has increased 6. 3% and 
 underground coal productivity (short tons 
 produced per 8-h shift) has increased 41% 
 over the same time span 05). Recently, 
 Spokes (6) found a correlation between 
 declining accident rates and increasing 
 productivity. However, this does not 
 necessarily mean that increasing the pro- 
 ductivity will definitely cause fewer ac- 
 cidents; there are many other factors 
 intertwined. 
 
 Examining the severity rates in figure 
 4 does not show any definite trends ex- 
 cept that both severity rates have errat- 
 ically increased over the 5-yr study 
 period. The overall severity rate 
 increased approximately 30%, and the non- 
 fatal severity rate increased less than 
 7%. This is a possible indication that 
 the seriousness of injuries resulting 
 from roof-rib fall accidents is on the 
 increase. 
 
 v> 
 
 1980 1981 1982 1983 1984 
 
 FIGURE 4.— U.S. roof-rib severity rates, 1980-84. A, number 
 of days lost (related to nonfatal accidents) per 200,000 
 employee-hours worked; 8, number of days lost and days 
 charged (related to fatal and permanent total disability ac- 
 cidents) per 200,000 employee-hours worked. 
 
STATE ACCIDENT RATES 
 
 Examination of equivalent accident 
 rates by State points out some regional 
 differences. Figures 5 and 6 show the 
 roof-rib fall accident rates for the top 
 10 underground coal producing States. 
 The following comparisons were drawn from 
 these figures and are compared with the 
 national rate. 
 
 Most noticeable in all of the bar 
 charts are the considerably higher roof- 
 rib fall accident rates in the Western 
 States of Utah and Colorado. On the 
 average, 1.8% of Utah's and Colorado's 
 work force was injured in roof-rib fall 
 accidents alone, as compared with the 
 national average of 0.9% (fig. 5c). This 
 may be due to several factors uniquely 
 associated with western coal mines. 
 These include abnormal seam characteris- 
 tics (deeper, thicker, and pitching 
 seams) and a less experienced work force. 
 In the east, the State of Virginia also 
 had fairly high accident rates (fig. 5). 
 However, there was a downward trend in 
 these higher accident rate States. 
 
 During the early part of the study, 
 Kentucky had the lowest roof-rib fall ac- 
 cident rates in the country but, by the 
 
 CHARACTERISTICS OF U.S. 
 
 The analysis of the compiled statistics 
 included detailed examination of roof-rib 
 accident characteristics related to the 
 time of occurrence, to specific mine at- 
 tributes, and to the accident victim. 
 Within each grouping, all available data 
 on each roof-rib accident were compiled 
 and evaluated. The data measured fre- 
 quency or rate of occurrence to emphasize 
 particular trends within each category. 
 
 SEASONAL PATTERNS 
 
 Data on the occurrence of roof-rib 
 falls with respect to time of year were 
 compiled to determine if seasonal 
 
 end of the study period, Tennessee, Ala- 
 bama, and Pennsylvania had equal or some- 
 what lower accident rates. Figure 5 in- 
 dicates that Kentucky's accident rates 
 were gradually approaching the national 
 rate, while rates of most other States 
 were on the decline. 
 
 Between these two extremes of accident 
 rates were Alabama, Illinois, Tennessee, 
 and West Virginia, which hovered around 
 the national rate (fig. 5). It is inter- 
 esting to note that West Virginia's acci- 
 dent rates closely followed the national 
 rates. Pennsylvania's and Ohio's acci- 
 dent rates, which were fairly high ini- 
 tially, dropped consistently after 1982 
 (fig. 5). 
 
 Severity rates in Tennessee, Virginia, 
 and the Western States were very high 
 initially, but then dropped or stabilized 
 (fig. 6). In a reverse situation, Ken- 
 tucky's rates were fairly low initially, 
 but then increased significantly above 
 the national average (fig. 6). Ken- 
 tucky's high overall severity rate in 
 1984 is due in part to the disproportion- 
 ate number of roof-rib fatalities (20 
 fatalities) that occurred that year. The 
 remaining States fall close to or below 
 the national severity rate. 
 
 ROOF-RIB FALL ACCIDENTS 
 
 patterns such as fluctuations in tempera- 
 ture, barometric pressure, and humidity 
 might affect the frequency of roof-rib 
 falls. Figure 7 displays the average 
 number of roof-rib fall accidents 
 (between 1980 and 1984) with respect to 
 month of accident. Data from 1981 were 
 omitted because it was a strike year and 
 monthly accident results were biased. 
 Also, roof-rib accidents that occurred in 
 mines of the Western United States (Utah, 
 Colorado, New Mexico, and Wyoming) were 
 omitted because these areas are mostly 
 arid climates and experience minimum 
 fluctuations in humidity. 
 
ALABAMA [ 
 
 COLORADO 
 
 ILLINOIS 
 
 KENTUCKY 
 
 OHIO 
 
 PENNSYLVANIA 
 
 TENNESSEE 
 
 UTAH 
 
 VIRGINIA 
 
 WEST VIRGINIA 
 
 UNITED STATES 
 
 ^2* 
 
 ^^^a^^^^^^^^^^^^^^w^ 
 
 M 
 
 ^^^y 
 
 ^m 
 
 **H, 
 
 MB 
 
 SS^^^^T 
 
 
 /////A 
 
 w 
 
 M& 
 
 r^ 
 
 ■ZZZ - 
 
 IP 
 
 E?SF 
 
 Irl'lir 
 
 KEY 
 IB 1980 
 E3I98I 
 □ 1982 
 
 E3I983 
 EUl 1984 
 
 riirnr 
 
 J I I L 
 
 * ':"': ~'_r "' 
 
 -\ 1 1 1 1 1 r 
 
 
 B 
 
 'M& 2 - 
 
 ^M 
 
 ^3 
 
 '///.'///A 
 
 m 
 
 rrrrnT 
 
 ^^^ 
 
 sss 
 
 'mm/MM '* * 
 
 ^_ "':"':" -^y 
 
 ^^^ 
 
 ^^ 
 
 
 J I I L 
 
 2357 
 
 W mr . V^ f^ 
 
 . 
 
 ^SE" 
 
 WW \\\\^\\\y\- w.wwl 
 
 
 1 1 r ii 1 1 imftm 
 
 a 
 
 2W 
 
 777Z7/f7T 
 
 tiffi- 
 
 \ \\\x\\ \\\V 
 
 SSS3 
 
 m wmim n * 
 
 2 4 6 8 10 12 14 04 0.81.2 1.6 2.024 28 3.2 3.6 
 
 RATE, accidents per — 
 
 0.01 
 
 0.02 
 
 0.03 
 
 MMst 
 
 200,000 
 employee -hours 
 
 Worker 
 
 FIGURE 5.— Roof-rib accident rates, by State. A, accidents per million short tons of underground coal mined; B, accidents per 
 200,000 employee-hours worked; C, accidents per average total number of underground workers. 
 
 Although it was not possible to normal- 
 ize the results based on employee-hours, 
 there did seem to be a seasonal trend. 
 The number of roof-rib accidents peaked 
 in the months of August through October 
 and then dropped off during the months of 
 
 November through February. This trend 
 may be due to an increase in coal produc- 
 tion during the late summer for winter 
 stockpiling, resulting in an increase in 
 underground exposure time and subsequent- 
 ly more accidents. Because monthly 
 
ALABAMA 
 
 COLORADO 
 
 ,200 
 
 RATE, days lost per RATE, days lost and 
 200,000 charged per 200,000 
 
 employee-hours employee -hours 
 
 FIGURE 6.— Severity rates for roof-rib accidents, by State. A, days lost (related to nonfatal accidents) per 
 200,000 employee-hours worked; B, days lost and days charged (related to fatal and permanent total disability ac- 
 cidents) per 200,000 employee-hours worked. 
 
 
50 
 
 y>* *e* ^ ^ ^ y»* ^ ^ <b& 0^ ^ O eC 
 MONTH 
 
 FIGURE 7.— Average number of roof-rib accidents with 
 respect to month of occurrences (excludes Western States). 
 
 production figures are unavailable, 
 this speculation could not be verified. 
 
 However, a similar seasonal trend of 
 roof falls was reported by Stateham and 
 Radcliffe (_7), who found that humidity 
 has a strong influence on roof fall oc- 
 currence rates. Using cubic regression 
 techniques, they correlated humidity and 
 roof fall statistics over a 3-yr period 
 and found that their best-fit curves al- 
 most coincided in sinusoidal cycles. The 
 roof fall occurrence rate curve followed 
 the humidity curve by about 14 days. 
 Their results indicated that the proba- 
 bility of a roof fall is greatest in Au- 
 gust and least in February. Their study 
 also indicated that the barometric pres- 
 sure was not related to roof falls. 
 
 Another study by Haynes (8) found that 
 the effect of temperatures and tempera- 
 ture changes on rock around mine openings 
 has a negligible effect on roof stabil- 
 ity. Consequently, it appears that the 
 only climate condition that may play a 
 part in the occurrence of roof-rib falls 
 is humidity. 
 
 DAILY PATTERNS 
 
 Another time-related parameter focuses 
 on the approximate hour at which the 
 roof-rib fall accident occurred. Figure 
 8 shows the number (5-yr average) of 
 roof-rib accidents within 30-min inter- 
 vals during a 24-h period. The peak num- 
 ber of accidents occurred between 10:00 
 and 10:30 a.m., 1:00 and 1:30 p.m., and 
 6:00 and 6:30 p.m. The 10:00-10:30 a.m. 
 peak may coincide with the ap-proximate 
 time of day miners are in full operation 
 at the face; however, by noon the number 
 of accidents has dropped by half, 
 
 i i i | i I i | ■ i i | r i i | i ' ' 1 ' ' ' I ' ' ' I ' ' ' I ' ' ' I ' ' ' 1 ' ' ' I 
 
 12 2 4 6 8 10 12 2 4 6 8 10 12 
 
 a.m. noon ►p.m. 
 
 30-MIN TIME INTERVALS 
 
 FIGURE 8.— Average number of roof-rib accident occur- 
 rences with respect to time of day. 
 
 probably because of the lunch break. The 
 1:00-1:30 p.m. peak may signify full 
 operation after lunch, and the 6:00-6:30 
 p.m. peak may be due to the evening shift 
 in full operation. These peaks may indi- 
 cate a pattern of more accidents occurr- 
 ing in the second and third hour into the 
 respective shifts and also after the 
 lunch break. 
 
 Theodore Barry and Associates (_9) found 
 a similar pattern in a study conducted in 
 the late 1960's. This study speculated 
 that the higher number of roof falls in 
 this period was due to lack of adequate 
 testing of the roof at the start of a 
 shift, since the miners, having just 
 started their daily tasks, were not fo- 
 cusing enough attention on the behavior 
 of the roof. Figure 8 shows a low number 
 of roof-rib accidents at the start and 
 end of each shift, probably because the 
 workers were in transition and away from 
 the face areas. These accident trends 
 can probably be related more to human 
 factors than to the occurrence of roof- 
 rib falls. 
 
 MINE ATTRIBUTES 
 
 According to the Department of Energy 
 (10), there were over 1, 760 underground 
 coal mines operating in the United States 
 in 1984. Although no two mines are iden- 
 tical, it was thought that common mine 
 characteristics such as seam height, mine 
 size, underground location mining method, 
 etc., may be associated with the frequen- 
 cy of roof-rib fall accidents. 
 
10 
 
 Seam Height 
 
 One of the obvious mine characteristics 
 that was considered was the seam height. 
 A scan of Figure 9A shows that roof-rib 
 accidents occur at a higher rate for thin 
 seams (<36 in) and for very thick seams 
 (>120 in). Examination of the fatality 
 rates (fig. 9B) also shows higher rates 
 for thin seams; however, the higher rates 
 do not extend to the thick seams. Figure 
 9 shows that accident and fatality rates 
 for mines with intermediate seam thick- 
 nesses (37-120 in) are fairly low. Data 
 for these figures cover only the 1983-84 
 period because the seam height was incon- 
 sistently reported for the other years. 
 
 The somewhat higher accident rates for 
 thick-seam mines, although not reflected 
 in the fatality rates, may possibly be 
 due to the higher roof, which allows any 
 
 falling material to gain more velocity 
 and thereby cause more serious injuries. 
 Possibly, the more extensive use of 
 canopies on mining equipment in thicker 
 seams may limit the occurrences of a fa- 
 tal roof-rib accidents. 
 
 The higher accident rates for thinner 
 seams (<36 in) may be due, in part, to 
 the extremely confined work area. The 
 low head room makes assessment of the 
 roof difficult and inhibits escape from 
 an impending roof fall. The thinner 
 seams also limit the type of mining that 
 can be used; for example, the use of 
 longwall mining, which may provide better 
 protection from roof-rib falls, is pre- 
 cluded. Another possible explanation 
 given by MSHA (11) is the lack of cabs 
 and canopies on low-coal mining face 
 equipment, resulting in less protection 
 and a higher frequency of accidents due 
 
 <30 
 31-36 
 37-42 
 43-48 
 
 £ 49-54 
 o 
 
 uj 55-60 
 
 x 
 
 mmmmwm 
 
 f////////////777. 
 
 < 
 
 UJ 
 CO 
 
 61-72 
 
 73-84 
 
 85-96 
 
 97-120 
 
 >I20 
 
 w/mmwmm ^ 
 
 m 
 
 a 
 
 Miffl l N I 
 
 WMZMML 
 
 TTTI 
 
 m 
 
 ? 
 
 TlTlTITITiriTII' 
 
 » 
 
 KEY 
 
 E2JI983 
 DID 1984 
 
 lllllllllllllllllll 
 
 mnmnmm)wnnun))))\ 
 
 n 
 
 F 
 
 0.5 
 
 1.0 
 
 1.5 
 
 RATE, accidents per 
 200,000 employee -hours 
 
 nwmm »»>«* 
 
 B 
 
 HIIIIIIIIIIIIIIIIIT 
 
 mr 
 
 (tminiuiK 
 
 m 
 
 ZZL 
 
 TTTTI 
 
 b 
 
 b 
 
 ? 
 
 2.0 
 
 mm 
 
 0.05 0.10 
 
 0.15 
 
 0.20 0.25 0.30 
 
 RATE, fatal accidents per 
 200,000 employee-hours 
 
 FIGURE 9.— Roof-rib accident rates with respect to seam height of the mine. A, fatal and nonfatal roof-rib accidents per 200,000 
 employee-hours worked; S, only fatal roof-rib accidents per 200,000 employee-hours worked. 
 
11 
 
 due to roof falls. The higher accident 
 rate may not be totally attributed to the 
 seam height, it may also be interrelated 
 with other factors such as the large num- 
 ber of low-coal mines that are small 
 mines. And as the succeeding section ex- 
 plains, smaller coal mines have higher 
 roof-rib accident rates. 
 
 Mine Size 
 
 The size of the mine was reviewed on 
 the assumption that larger mines have 
 larger technical staffs and more capital 
 to deal with ground control problems. 
 Small mines have a minimal technical 
 staff (if any), little capital, and are 
 sometimes located in unusual seams with 
 difficult ground control problems. Con- 
 sequently, the smaller mines may have a 
 higher accident rate, as documented in 
 accident studies conducted by the 
 National Academy of Science and National 
 
 Research Council (12-13). Their studies, 
 which included all types of accidents, 
 found a strong correlation between mine 
 size and fatal injuries. 
 
 Since the size of a mine can be quanti- 
 fied by the magnitude of its work force, 
 this Bureau study normalized all roof-rib 
 accidents and fatal accidents based on 
 the annual average total number of em- 
 ployees (fig. 10). Examination of the 
 accident rates (fatal and nonfatal) in 
 figure 10A shows that, initially, inter- 
 mediate-size mines (51 to 150 employees) 
 had high accident rates, but by 1983 
 these rates had dropped considerably. 
 Over the same period, accident rates in 
 small mines (1 to 20 employees) increased 
 above all the other mine size categories. 
 Even more pronounced is the fatal acci- 
 dent rate (fig. 105), which shows that 
 small mines have a significantly higher 
 fatality rate than all other groups. It 
 is interesting to note from both graphs 
 
 Id 
 
 l±J 
 
 5 
 
 -J 
 Q. 
 
 LLi 
 
 b 
 
 I - 20 t 
 
 T 
 
 wm<m<m 
 
 n 
 
 2i -so ; 
 
 m<mm\ 
 
 51-150 
 
 mm<m 
 
 KEY 
 1980 
 
 ^1981 
 □ 1982 
 ^1983 
 ^1984 
 
 & 
 
 ' B 
 
 wmmm 
 
 0.5 1.0 1.5 2.0 2.5 
 
 RATE, accidents per 
 200,000 employee -hours 
 
 0.04 0.08 0.12 0.16 0.20 
 RATE, fatal accidents per 
 200,000 employee -hours 
 
 FIGURE 10.— Roof-rib accidents rates with respect to mine size based on the average number of mine employees. A, fatal and 
 nonfatal roof-rib accidents per 200,000 employee-hours worked; B, only fatal roof-rib accidents per 200,000 employee-hours 
 worked. 
 
12 
 
 that very large mines (greater than 250 
 employees) have the lowest roof -rib acci- 
 dent rates of all categories, especially 
 for fatal accidents. 
 
 Underground Location 
 
 The location of the roof-rib fall acci- 
 dent was another factor studied. Fig- 
 ure 11 shows that approximately 61% of 
 roof-rib fall accidents occur at the 
 working face, 12% at an intersection, and 
 the remaining portion at various other or 
 undefined locations. These results seem 
 reasonable since most miners work at the 
 face. At the face, the miners are often 
 under temporarily supported or unsup- 
 ported roof, where there is a much great- 
 er risk of a roof fall. No trends from 
 year to year were detectable with respect 
 to location. 
 
 Mining Method 
 
 Further examination of mine attributes 
 involved the mining method employed at 
 the location where the accident occurred. 
 Longwall mining is becoming a popular 
 method of mining, which is evident by a 
 100% increase in the number of working 
 longwall faces in the United States be- 
 tween 1974 and 1986 (14-15). The rapid 
 growth in use of the longwall method may 
 be attributed to a higher recovery rate 
 and safer working climate, since the min- 
 ers are almost always under the protec- 
 tion of the longwall shields. 
 
 Figure 12 is a compilation of the num- 
 ber of roof-rib accidents based on the 
 mining method used at the section where 
 the accident occurred. The bar chart 
 shows accidents decreasing for the con- 
 ventional and continuous mining methods, 
 with the number of longwall accidents in- 
 creasing slightly with time. This slight 
 increase is most likely due to the 
 increased number (approximately 59 pan- 
 els) of longwall panels started in the 
 5-yr period. 
 
 For a more accurate assessment of acci- 
 dent trends, the data needed to be nor- 
 malized based on annual tonnage per min- 
 ing method. Table 1 shows the accident 
 rates of the longwall method versus all 
 other methods (conventional, continuous, 
 
 Intersections 
 12.5% 
 
 Face 
 61.3% 
 
 Shaft, shop, 
 slope 
 3% 
 
 Other or undefined 
 23% 
 
 FIGURE 11.— Roof-rib accident locations. 
 
 KEY 
 
 MI980 
 
 □ I98I 
 
 □ 1982 
 
 □ 1983 
 ^1984 
 
 Continuous 
 
 Conventional 
 MINING METHOD 
 
 ^^ZEfi 
 
 Longwall 
 
 FIGURE 12.— Frequency of roof-rib accidents with respect 
 to type of mining method. 
 
 and hand loading) grouped together and 
 normalized with production figures from 
 1978 and 1983, since these are the only 
 years for which longwall production fig- 
 ures were published ( 10 , 16). However, 
 these rates are somewhat biased since 
 longwalls are considerably more produc- 
 tive than other mining methods and the 
 number of accidents that the longwall 
 rate is based upon is considerably less 
 than the number of accidents for the 
 other rate, which may affect the 
 accuracy. 
 
 Both longwall and the other mining me- 
 thods show a substantial accident rate 
 drop over the 5-yr time span (from 3.5 to 
 1.48 accidents per million short tons for 
 longwalling and 4. 88 to 2. 98 accidents 
 for all other methods) reflecting the 
 overall drop in roof-rib accidents. Yet, 
 the percentage difference in roof-rib ac- 
 cident rates between the longwall method 
 and all other methods widened from 28% in 
 1978 to 50% in 1983 (table 1). 
 
 Although these data are somewhat 
 biased, they may indicate that longwalls 
 provide a safer working environment, as 
 was found in an equivalent accident study 
 conducted by Peake (17). Peake's study, 
 
13 
 
 TABLE 1. - Roof-rib accident rates, longwall versus all other 
 mining methods 
 
 Production, st: 
 
 Longwall 
 
 Other methods 
 
 Number of roof -rib accidents: 
 
 Longwall 
 
 Other methods 
 
 Accident rate: ' 
 
 Longwall 
 
 Other methods 
 
 Difference 2 %. 
 
 1978 
 
 1983 
 
 11,981,000 
 
 47, 
 
 257 
 
 ,000 
 
 217,094,592 
 
 244, 
 
 885 
 
 ,378 
 
 42 
 
 
 
 70 
 
 1,061 
 
 
 
 731 
 
 3.50 
 
 
 
 1.48 
 
 4.88 
 
 
 
 2.98 
 
 28 
 
 
 
 50 
 
 Accidents per million short tons mined. 
 
 Percentage by which roof-rib accident frequency for all 
 other mining methods exceeded that for longwall mining in the 
 United States. 
 
 conducted in 1985, examined all types of 
 accidents for 13 longwall and nonlongwall 
 working faces and the data were normal- 
 ized based on employee-hours worked in- 
 stead of production. Peake's study found 
 that the nonlongwall raining method acci- 
 dent rate exceeded the longwall rate by 
 53%. 
 
 THE ACCIDENT VICTIM 
 
 One of the more important concerns in 
 compiling these statistics was to evalu- 
 ate the characteristics and effects that 
 a roof-rib accident has on the victims 
 and their families. Various factors 
 evaluated include the victim's activity 
 at the time of the roof-rib accident, 
 the number of lost workdays, and the type 
 of injury resulting from the accident. 
 
 Worker Activity 
 
 The types of activities that the vic- 
 tims were pursuing at the moment the 
 roof-rib fall occurred were evaluated. 
 Tables 2 and 3 show the top 10 worker ac- 
 tivities, with the highest accident and 
 fatality rates based on employee-hours 
 worked (5-yr average). Handling sup- 
 plies, barring down the roof-rib, and 
 roof bolting had the highest accident 
 rates (table 2). Examination of the fa- 
 tality rates in table 3 shows almost the 
 same worker activities as table 2 except 
 in a different order, possibly indicating 
 
 TABLE 2. - Worker activity roof-rib 
 accident rates, averages for 
 1980-84, accidents per 200,000 
 employee-hours 
 
 Rank 
 
 1.. 
 2.. 
 3.. 
 4.. 
 5.. 
 6.. 
 7.. 
 8.. 
 9.. 
 10. 
 
 Worker activity 
 
 Handling supplies or material 
 
 Bar down 
 
 Roof bolter, other 
 
 Set, remove, relocate props.. 
 
 Continuous miner 
 
 Walking, running 
 
 Idle 
 
 Machine maintenance, repair.. 
 
 Timbering 
 
 Move power cable 
 
 0.088 
 .081 
 .070 
 .061 
 .060 
 .050 
 .050 
 .045 
 .043 
 .035 
 
 TABLE 3. - Worker activity roof-rib 
 fatality rates, averages for 
 1980-84, fatalities per 200,000 
 employee-hours 
 
 Rank 
 
 1... 
 2... 
 
 ■J * • >> 
 
 4... 
 5... 
 6. • • 
 7... 
 8. . . 
 9... 
 10.. 
 
 Worker Activity 
 
 Continuous miner 
 
 Timbering 
 
 Observe operations 
 
 Roof bolter, inserting bolt 
 
 Handling supplies 
 
 Supervise 
 
 Set, remove, relocate props 
 
 Roof bolter, other 
 
 Bar down 
 
 Walking, running 
 
 0.0035 
 .0025 
 .0022 
 .0022 
 .0021 
 .0021 
 .0019 
 .0018 
 .0017 
 .0014 
 
 which activities are more prone to result 
 in a fatal accident. Operating the 
 
14 
 
 continuous miner, timbering, and observ- 
 ing operations had the highest fatality 
 rates. Although many of the activities 
 listed in table 3 correspond to activi- 
 ties that usually occur at the working 
 face, several of the activities were 
 merely observing operations, supervising, 
 walking, or idle time. These data empha- 
 size the randomness at which a roof-rib 
 fall can occur. 
 
 Lost Workdays 
 
 Most of the attention associated with 
 roof-rib accidents is focused on the fa- 
 talities, and rightly so. However, out 
 of the average 1, 100 roof-rib accidents 
 that occur each year, less than 4% of 
 these accidents are fatalities. Whereas, 
 over 50% of these accidents are severe 
 enough to result in 10 or more lost work- 
 days, as shown in table 4, which gives a 
 cumulative percentage of days lost. 
 
 MSHA defines days lost as the number of 
 full calendar days that the injured em- 
 ployee is unable to work as a result of a 
 temporary disability (3). This does not 
 include lost workdays from an accident 
 resulting in a permanent total disabil- 
 ity. Consequently, the statistics offer 
 a relative indication of the seriousness 
 of roof-rib accidents in terms of the 
 number of lost workdays. 
 
 Data from figure 13 indicat-e that the 
 average length of time away from work be- 
 cause of a roof-rib fall accident has in- 
 creased from 35 to 44 days over the 5-yr 
 study period. This indicates that the 
 victim is requiring a longer recuperation 
 period, and one reason may be that roof- 
 rib accidents are becoming more severe, 
 as was noted with the severity rates. A 
 
 TABLE 4. - Distribution of roof-rib 
 accidents with respect to days 
 lost, averages for 1980-84 
 
 Days lost 
 
 >499 
 
 249 to 499. 
 143 to 249. 
 99 to 143.. 
 88 to 99... 
 66 to 88... 
 44 to 66. .. 
 22 to 44... 
 10 to 22... 
 
 4 to 10 
 
 1 to 4 
 
 
 
 1 
 
 Total. 
 
 Av No. of 
 accidents 
 
 2 
 
 19 
 
 42 
 
 42 
 
 18 
 
 49 
 
 90 
 
 155 
 
 160 
 
 176 
 
 157 
 
 148 
 
 40 
 
 1,098 
 
 Cumulative 
 percentage 
 
 0.2 
 
 1.9 
 
 5.7 
 
 9.5 
 
 11.2 
 
 15.6 
 
 23.9 
 
 37.9 
 
 52.5 
 
 68.5 
 
 82.9 
 
 96.3 
 
 100.0 
 
 NAp 
 
 NAp Not applicable. 
 
 'Fatal and other accidents resulting in 
 permanent total disability. 
 
 tally of the total workdays lost between 
 1980 and 1984 yields almost 200,000 days 
 or 1.6 million employee-hours lost be- 
 cause of roof-rib accidents alone (These 
 figures do not include workdays lost from 
 an accident resulting in a permanent 
 total disability. ) 
 
 Type of Injury 
 
 Analysis of the types of injuries sus- 
 tained by roof-rib fall victims shows 
 that 17% of the victims suffer injuries 
 to multiple parts of their bodies. Mul- 
 tiple injuries are most often linked with 
 severe or fatal injuries. Following mul- 
 tiple injuries, the most frequent 
 injuries are to the back and the 
 
 1980 1981 1982 1983 
 
 FIGURE 13.— Average number of lost workdays due to roof-rib accidents. 
 
 I984 
 
15 
 
 extremities (finger, foot, hand, and 
 leg). These injuries are usually less 
 sever than multiple injuries. The parts 
 of the body injured in roof-rib accidents 
 seem fairly well represented because fall 
 material injures susceptible areas of 
 the body such as the back, extremities, 
 etc. In many instances, injuries affect 
 more than one area. 
 
 Roof-rib accidents not only affect the 
 victim physically but also financially. 
 This introduces a new factor: What eco- 
 nomic effects do roof-rib accidents have 
 on their victims, as well as on the 
 mining industry? 
 
 COST ANALYSIS OF ROOF-RIB FALL ACCIDENTS 
 
 To evaluate the economic effects that 
 roof-rib fall accidents cause, the com- 
 puter software package ACIM (accident 
 cost indicator model) was utilized. This 
 cost analysis program, developed under 
 contract for the Bureau, was mathemati- 
 cally modeled to incorporate information 
 gathered from mine inspection offices, 
 workmen's compensation agencies, insur- 
 ance companies, and major medical cen- 
 ters. Output from the program estimates 
 the tangible costs, both total cost and 
 cost per accident, for a specified type 
 and degree of accident. The costs are 
 broken down into the following categories 
 ( 18-19 ) : 
 
 1. The cost of the accident to the 
 mining industry. — This includes the cost 
 of cleanup after the accident and associ- 
 ated production losses, the cost of State 
 Worker Compensation Benefits, the cost of 
 the medical treatment, the cost of the 
 union death-disability benefits, and the 
 cost of the mining industry's investiga- 
 tion of the accident. 
 
 2. The cost of the accident to the 
 victim and victim's family. — Cost due to 
 lost wages during recuperation or the 
 lost wages for the remaining portion of 
 the victim's career if the accident re- 
 sults in a fatality or permanent (total) 
 disability. Lost wages are adjusted to 
 account for compensation wages received 
 from benefits. 
 
 3. The cost of the accident to the 
 public sector. — This includes the cost of 
 benefits (Federal Social Security) paid 
 out to the victim and victim's family and 
 the cost of the public investigation con- 
 ducted by MSHA. 
 
 These costs are based on cost data for 
 a particular year and are not adjusted 
 for inflation unless otherwise stated. 
 Also excluded are costs of lawsuits; 
 costs of hiring and training replacement 
 workers; costs of resarch to prevent or 
 reduce accidents; and most costs related 
 to lost profit, sales, or equipment idled 
 by an accident (18). 
 
 It should be emphasized that the ACIM 
 program does not use the actual costs 
 associated with each accident but rather 
 a random generator to approximate the ac- 
 cident costs. Therefore, these costs 
 should be used cautiously since they are 
 only estimated costs of roof-rib acci- 
 dents. This is especially true for fatal 
 accidents, where no true cost can be 
 placed on the loss of human life. 
 
 TOTAL ACCIDENT COSTS 
 
 The software program approximates the 
 total cost of all roof-rib accidents be- 
 tween 1980 and 1984 at $215 million or 
 about 27% of the cost of all underground 
 accidents. A breakdown of the groups of 
 individuals economically affected by 
 roof-rib accidents (averaged 1980-84) is 
 shown in the pie chart in figure 14. The 
 major costs of these accidents between 
 1980 and 1984 are carried by the mining 
 industry (approximately 47% or $100 mil- 
 lion), the victim and victim's family 
 (approximately 32%, or $70 million), and 
 public agencies (approximately 21% or $45 
 million). Also shown in figure 14 is the 
 breakdown of the cost percentage involved 
 within each grouping. The major roof -rib 
 accident costs result from production 
 losses ($37 million), lost wages ($70 
 
16 
 
 MINING 
 INDUSTRY iProduc 
 COST 
 
 Workmen s compensation, 
 
 39.7%, and investigation 
 
 cost, 0.3% 
 
 Union 
 
 benefits 
 10.7 % 
 
 nvestigation 
 cost 2% 
 
 FAMILY COST 
 
 PUBLIC COST 
 
 FIGURE 14.— Pie chart of roof-rib accident cost percent- 
 ages, averages for 1980-84. 
 
 million), and benefits, such as Social 
 Security, union, and worker's compensa- 
 tion ($95 million). 
 
 COST PER ACCIDENT 
 
 To obtain a more accurate approximation 
 of accident costs, the program normalizes 
 
 the data in the form of cost per acci- 
 dent. Table 5 illustrates the wide gap 
 that exists between the cost per accident 
 of roof-rib fall accidents versus all 
 other accident types. In 1984, the cost 
 per roof-rib fall accident was estimated 
 at $52,000 while the cost per accident of 
 all other underground accidents was esti- 
 mated at $16, 600. 
 
 As table 5 shows, roof-rib accidents 
 are over three times as costly as all 
 other underground accidents. Specifical- 
 ly, these large roof-rib costs can be 
 traced to the cost of production losses 
 (6.2 times more costly per accident), the 
 cost of all benefits (1.3-4.1 times more 
 costly per accident), and the cost of 
 lost wages (3. 9 times more costly per 
 accident). The higher production loss 
 costs may be related to a longer shutdown 
 period of the mine because of cleanup of 
 the fallen material and resupport of the 
 roof-rib fall area. The higher union, 
 Federal, and workmen's compensation bene- 
 fit costs and wage losses may be due to 
 the greater severity of injuries caused 
 by roof-rib fall accidents requiring a 
 longer recuperation. Thus, the roof-rib 
 accident victim loses more wages and sub- 
 sequently increases the cost of the bene- 
 fits that are paid by the mining industry 
 and public sectors. Consequently, these 
 large increases in cost per accident sub- 
 stantiates the negative impact that 
 
 TABLE 5. -Comparison of roof-rib accident costs and all other types of 
 accidents for 1984, cost per accident 
 
 
 Roof-rib 
 accidents 
 
 All other mining 
 accidents 
 
 Ratio 
 
 Mining industry: 
 
 $6,929 
 
 10,917 
 
 774 
 
 2, 144 
 
 63 
 
 $5,322 
 
 1,746 
 
 611 
 
 514 
 
 13 
 
 1.30 
 
 
 6.25 
 
 
 1.27 
 
 
 4. 17 
 
 
 4.85 
 
 
 20,827 
 
 8,206 
 
 2.54 
 
 Family: 
 
 Lost wages (total cost to family).. 
 
 21,324 
 
 5,488 
 
 3.88 
 
 Public: 
 
 9,714 
 231 
 
 2,860 
 45 
 
 3.40 
 
 
 5.13 
 
 
 9,945 
 
 2,905 
 
 3.42 
 
 
 52,096 
 
 16,599 
 
 3.14 
 
 Benefits. 
 
17 
 
 roof-rib accidents have on various sec- 
 tors of the economy. 
 
 To evaluate cost trends over the 5-yr 
 period, the ACIM program was modified 
 such that all the costs were in 1983 dol- 
 lars. This eliminated the effects of in- 
 flation over the time period and allowed 
 the data to be equitably compared. Fig- 
 ure 15i4 shows that roof -rib accidents as 
 well as all other types of accidents have 
 erratically increased in cost (approxi- 
 mately 30%) over the time period. There- 
 fore, all types of accidents are becoming 
 increasingly more expensive, even with 
 the effects of inflation eliminated. 
 
 COMPARISON OF ACCIDENT COSTS BASED ON 
 DEGREE OF INJURY 
 
 Figures 15B and 15C display the approx- 
 imate cost ranges of roof-rib accidents 
 with varying degrees of injury. Figures 
 15B and 15C illustrates that the degree 
 
 40 
 30 
 20 
 
 o 
 
 10 
 
 o 
 
 1,400 
 
 ro 
 O 
 
 1,200 
 
 
 1,000 
 
 K 
 
 
 Z 
 
 800 
 
 UJ 
 
 
 Q 
 
 600 
 
 O 
 
 
 O 
 
 400 
 
 < 
 
 
 
 200 
 
 a: 
 
 8 
 
 LU 
 
 
 Q. 
 
 
 1- 
 
 6 
 
 cn 
 
 
 o 
 
 
 o 
 
 A 
 
 KEY 
 
 Roof- rib accidents 
 All other accidents 
 
 KEY 
 
 Fatal 
 
 Permanently disabled 
 
 4 
 
 2 - 
 
 KEY 
 
 1 Injured with days lost 
 Injured with no days lost 
 
 1980 
 
 1981 
 
 1982 
 
 1983 
 
 1984 
 
 FIGURE 15.— Comparison of costs per accident. A, roof-rib 
 accidents versus all other types of accidents (adjusted for in- 
 flation); B, fatal roof-rib accidents versus permanent partial or 
 total disabling accidents; C, roof-rib accidents resulting in 
 days lost injuries versus no days lost injuries. 
 
 of injury dramatically dictates the cost 
 of a roof-rib accident (based on cost per 
 accident in 1983 dollars). For instance, 
 the cost per fatal accident (degree 1) is 
 around $1 million per accident, but the 
 cost of an accident resulting in a perma- 
 nent partial or total disability (degree 
 2) is around $300,000; while the cost of 
 an accident resulting in lost workdays is 
 close to $6,000, and the cost of an acci- 
 dent with no lost workdays is about 
 $500. 
 
 For actual cost comparisons of various 
 degrees of injury, roof-rib accident 
 costs for 1983 were examined as shown in 
 table 6. Comparison of the fatal acci- 
 dents (degree 1) versus accidents result- 
 ing in permanent disabilities (degree 2), 
 which are the two most costly types of 
 accidents, yields some interesting infor- 
 mation. The estimated cost per accident 
 resulting in a permanent disability is 
 about 40% of the cost of a fatal acci- 
 dent. The lower costs of degree 2 acci- 
 dents result from considerably lower pro- 
 duction losses, lost wages, and benefit 
 costs (table 6). Whenever a fatal acci- 
 dent occurs, it forces the whole mine to 
 shutdown for one or more workdays, while 
 with degree 2 accidents only the affected 
 mine section may be closed. Consequent- 
 ly, degree 2 accidents have lower produc- 
 tion losses than occur with fatal 
 accidents. 
 
 Another comparison was made from table 
 6, contrasting fatal accidents (degree 1) 
 with accidents resulting in injuries with 
 lost workdays (degrees 3-4) and injuries 
 with no workdays lost (degrees 5-6), 
 These comparisons are even more note- 
 worthy. The degree 1 accident cost aver- 
 aged $955, 000 per accident while the de- 
 grees 3 and 4 accident cost averaged 
 $6,123 per accident (150 times less than 
 degree 1 accident costs) and the degrees 
 5 and 6 accident cost averaged $517 per 
 accident (1,800 times less than degree 1 
 accident costs). 
 
 The lower accident cost of degrees 3 
 and 4 is due to zero production losses 
 (more than likely a minuscule amount of 
 production losses did occur), lower bene- 
 fit costs, and less lost wages. With a 
 less severe accident injury, the mine is 
 not closed and therefore production 
 
18 
 
 TABLE 6. - Comparison of costs per roof-rib accident based on degree of injury 
 for 1983, cost per accident 
 
 Cost factors 
 
 Degree 1: 
 Fatal 
 
 Degree 2: 
 
 Permanently 
 
 disabled 
 
 Degrees 3-4: 
 injured with 
 lost workdays 
 
 Degrees 5-6: 
 
 Injured with no 
 
 lost workdays 
 
 Mining industry: 
 
 $126,668 
 
 229,302 
 
 
 
 55,117 
 
 1,640 
 
 $179,838 
 
 9,415 
 
 1,048 
 
 22,124 
 
 
 
 $3,023 
 
 
 
 825 
 
 32 
 
 
 
 
 
 
 
 $517 
 
 o 
 
 
 
 
 
 
 412,727 
 
 212,425 
 
 3,880 
 
 517 
 
 Family: 
 
 Lost wages (total family 
 
 263,953 
 
 85,023 
 
 1,940 
 
 
 
 Public: 
 
 Social Security benefits'. 
 
 273,301 
 5,079 
 
 103,329 
 
 
 303 
 
 
 
 
 
 
 
 
 278,380 
 
 103,329 
 
 303 
 
 
 
 Total cost per accident 
 
 955,060 
 
 400,777 
 
 6,123 
 
 317 
 
 Total accident cost 2 ... 
 
 21,966,400 
 
 3,606,998 
 
 3,968,407 
 
 62,535 
 
 Benefits. 
 2 Total estimated cost of roof -rib accidents for all degrees of injury in 1983 
 $29,604,340. 
 
 losses do not occur. Also, lost wages 
 and benefit costs drop because of less 
 severe injuries and shorter recuperation 
 time. Costs of accidents with injuries 
 of degrees 5 and 6 are reduced even fur- 
 ther, almost to zero, except for the med- 
 ical costs involved. Therefore, as the 
 severity of the injury decreases the cost 
 per accident decreases exponentially and 
 considerably reduces the financial burden 
 of the mining industry, the family, and 
 the public. 
 
 The total estimated cost of all roof- 
 rib accidents in 1983, which was the low- 
 est annual cost of the five years stud- 
 ied, was calculated at approximately, 
 $29,600,000 (table 6). Although these 
 costs are only estimates, they are a mon- 
 etary incentive for the mining industry 
 to provide a safer underground environ- 
 ment and to instill in its workers safe 
 working habits for reducing the risk of 
 roof-rib fall accidents. 
 
 SUMMARY 
 
 The effects of roof-rib accidents are 
 extensive, ranging from the economic loss 
 of equipment and production to the fatal 
 and nonfatal injuries that result in 
 lasting physical and financial impair- 
 ments suffered by the victims and their 
 families. Although roof-rib falls will 
 probably never be totally eliminated, the 
 statistics show that the following prob- 
 lem areas need further attention: 
 
 - High accident rates in Virginia, Colo- 
 rado, and Utah. 
 
 - Fairly high severity rates in Utah and 
 Kentucky. 
 
 - Increased risk of roof falls in August 
 through October because of higher 
 humidity. 
 
19 
 
 - More accidents seem to occur in the 
 second and third hour of each shift and 
 after the lunch break. 
 
 - High accident rates in mines with very 
 thin coal seams (<36 in) and with very 
 thick seams (>120 in). 
 
 - High fatality rates in mines with very 
 thin coal seams (<36 in). 
 
 - High fatality rates at small mines with 
 average annual work force of less than 
 20. 
 
 - Several worker activities seem more 
 prone to accidents, such as operating 
 continuous miners and roof bolting. 
 
 - Increasing number of days away from 
 work after a serious injury. 
 
 Underlining the effects of roof-rib 
 fall accidents are the accident costs, 
 which have been estimated at $52,000 per 
 roof-rib fall accident as opposed to 
 $16,600 for all other types of accidents. 
 Also, as the degree of injury resulting 
 from a roof -rib accident becomes 
 
 more severe, the cost of the accident 
 increases exponentially. These high 
 costs impact the mining industry, the in- 
 jured workers and their families, and in- 
 directly the public sector. 
 
 These compiled roof-rib statistics also 
 indicate the following encouraging areas 
 that should be maintained: 
 
 - U.S. roof -rib accident rates have con- 
 sistently dropped in the last 20 yr, 
 especially in recent years. 
 
 - A lower probability of accidents exists 
 in mines with coal seam thicknesses of 37 
 to 120 in. 
 
 - Mines employing more than 250 workers 
 have a lower probability of accidents. 
 
 - Longwall mining appears to be 
 mining method. 
 
 a safer 
 
 While these statistics may not be abso- 
 lute, they do offer a current profile of 
 U.S. roof and rib accidents and related 
 accident costs, which reinforces the need 
 for continued ground control research. 
 
 REFERENCES 
 
 1. Peng, S. S. Coal Mine Ground Con- 
 trol. Wiley, 1978, 200 pp. 
 
 2. Thrush, P. W. A Dictionary of 
 Mining, Mineral, and Related Terms. Bu- 
 Mines Spec. Publ. , 1968, p. 151. 
 
 3. U.S. Mine Safety and Health Admin- 
 istration (Dep. Labor). Summary of Se- 
 lected Injury Experience and Worktime for 
 the Mining Industry in the United States, 
 1931-1977. 1984, 69 pp. 
 
 4. Schlick, D. P., R. Peluso, and 
 K. Thirumalai. U.S. Coal Mining Acci- 
 dents and Seam Thicknesses. Paper in 
 Proceedings of Symposium on Thick Seam 
 Mining by Underground Methods (Queens- 
 land, Australia, 1976). Australasian 
 Inst. Min. and Metall. , Symp. Ser. 
 No. 14, Aug. 1976, pp. 61-74. 
 
 5. U.S. Mine Safety and Health Admin- 
 istration (Dep. Labor). Mine Injuries 
 and Worktime, Quarterly: Closeout Edi- 
 tions 1980-84. 119 pp. 
 
 6. Spokes, E. M. New Look at Under- 
 ground Coal Mine Safety. Min. Eng. (Lit- 
 
 tleton, CO), v. 
 pp. 266-270. 
 
 7. Stateham, R. 
 cliffe. Humidity: 
 Coal Mine Roof 
 
 26, No. 4, 1986, 
 
 M. , and D. E. Rad- 
 A Cyclic Effect in 
 Stability. BuMines 
 RI 8291, 1978, 19 pp. 
 
 8. Haynes, C. D. Effects of Tempera- 
 ture and Humidity Variations on Coal Mine 
 Roof Stability. Paper in Proceedings 
 of a Symposium on Underground Mining 
 (Louisville, KY, Oct. 21-23, 1975). 
 Natl. Coal Assoc. , Washington, DC, v. 2, 
 1975, pp. 120-126. 
 
 9. Theodore Barry and Associates. 
 Accident Prediction Investigation Study 
 (contract S0122023). BuMines OFR 38-73, 
 1972, 174 pp.: NTIS PB 221000. 
 
 10. U.S. Energy Information Adminis- 
 tration (Dep. Energy). Coal Production 
 1984. DOE/EIA-0118, 1985, 144 pp. 
 
 10788 
 
 95 
 
20 
 
 11. U.S. Mine Safety and Health Admin- 
 istration (Dep. Labor). Comparison of 
 Injury Hazards in Different Coal Seams 
 Heights. 1981, 22 pp. 
 
 12. National Academy of Sciences. Fa- 
 talities in Small Underground Coal Mines 
 (contract J010014-5). BuMines OFR 124- 
 83, 1983, 20 pp. 
 
 13. National Research Council. Toward 
 Safer Underground Coal Mines. Natl. 
 Acad. Press, 1982, 190 pp. 
 
 14. Loxley, T. E. , D. B. Lull, and 
 J. 0. Rasor. Self-Advancing Roof Sup- 
 ports for Longwall and Shortwall Mining, 
 October 1974 Census, U.S. Naval Surface 
 Weapons Center, Apr. 1975, 113 pp. 
 
 15. Sprouls, M. W. Longwall Census 
 '84. Coal Min. and Process., v. 21, Dec. 
 1984, pp. 39-53. 
 
 16. U.S. Energy Information Adminis- 
 tration (Dep. Energy). Bituminous Coal 
 
 and Lignite Productions and Mine Opera- 
 tions. DOE/EIA-0118(78), 1980, 80 pp. 
 
 17. Peake, C. V. Longwall Output Con- 
 tinues to Rise. Coal Age, v. 91, No. 8, 
 1986, pp. 58-60. 
 
 18. Chi, D. N., and D. G. Di Canio. 
 Mine Accident Cost Data Bases and Some 
 Implications. Pres. at 1983 Am. Min. 
 Congr. Min. Conv. , San Francisco, CA, 
 Sept. 11-14, 1983, 5 pp., available from 
 D.N. Chi, BuMines, Pittsburgh, PA. 
 
 19. Di Canio, D. G. , and A. H. Nakata. 
 Accident Cost Indicator Model To Estimate 
 Costs to Industry and Society From Work- 
 Related Injuries and Deaths in Under- 
 ground Coal Mining, Volume I. Develop- 
 ment and Application of Cost Model 
 (contract J0255031, FMC Corp.). BuMines 
 OFR 39(l)-77, 1976, 202 pp.; NTIS PB 264 
 438. 
 
 U.S. GOVERNMENT PRINTING OFFICE 1987 605 017 60091 
 
 INT.-BU.0F MINES,PGH. ,PA. 28541 
 
U.S. Department of the Interior 
 Bureau of Mines— Prod, and Distr. 
 Cochrans Mill Road 
 P.O. Box 18070 
 Pittsburgh. Pa. 15236 
 
 OFFICIAL BUSINESS 
 
 PENALTY FOR PRIVATE USE. S300 
 
 [ ] Do not wish to receive this 
 material, please remove 
 from your mailing list* 
 
 j ~] Address change* Please 
 correct as indicated* 
 
 AN EQUAL OPPORTUNITY EMPLOYER 
 

 
 w 
 
 ♦♦ v % 
 
 V 
 
 
 
 v t * 
 
 7 » A V< ^ 
 
 
 u-rf 1 %«^V. '++<$ :^&. ^ :£&&:. "++#- s^mt\ *hf :2§f^»'. "W 
 
 * *> !^L'. 
 
 <J> " . « ° V 
 
 
 
 v^ 
 
 J ^ 
 
 
 
 
 ^ 
 ^ 
 
 
 
 
 
 
 ,« *5^ ^A 
 
 1- V *v» - 
 
 » y ^ 
 
 • a? r«_ » 
 
 9 ^^'. > 
 
 v > 
 

 p <?* £7 * Mil * «S* <V 
 
 :t«\a 
 
 *°* 
 
 <^ * 
 
 /,1^%\ G *.^- * /\v^\ c° 
 
 ./ «^ v.< c .\sa*. o *^ ^ 
 
 
 *° 
 
 "!rl^*. «> 
 
 
 * 4/ V«» > 
 
 
 
 •bV 
 
 o*- '^ *~ -v^^ar- ^^ 
 
 a^.\. 
 
 & 
 
 ^ 
 
 v^y ... X^/ °^° 
 
 
 :- 'W •« 
 
 
 
 V*# 
 
 
 
 
 "• V'MV 
 
 
 •w :3a^»'- **o« ;oS '»>.*' /«* '**„< 
 
 
 VV 
 
 . • j^5^s™»,*- o 
 
 
 $ 
 
 °* 
 
 *bv 
 
 
 1 *'*/ 1 L'* ^ 
 
 ,0 V ^b, "-"^TIT 4 * A 
 
 <* 
 
 -- 
 
 ^ * e « o 9 . -^ 
 
 
 
 *. % 
 
 M <y <<* o 
 
 
 >~ *> v , » • • 
 
 
 ;- ^ n < o^ 
 
 
 %p^. 
 
 
 ° - / -VS 
 
 «>. * e n o ° ^ v 
 
 » V '♦ V A t ■ - "» i*v a 
 
 . . » ^ ^ V '^K^ A ^ ''fT?- 4 .6*" V 7 ^ 7 '' ^ 
 
 v X^ v SMk- ~\<f : »Bfv ^ %/' 
 
 \