13
creating a crosscut using the first method led to instability,
while the second method produced a more stable crosscut.
As illustrated in Figure 18, the yield pattern in the immedi-
ate roof from the FLAC3D model for the second method is
significantly less than that of the first method. Therefore, it
is recommended to develop new crosscuts using the second
method when the roof is subjected to high horizontal stress.
Based on field observation, when the first method was
used to create a crosscut, the crosscut was stable during
the first three cuts however, when the last cut was done,
the crosscut becomes unstable and cutter failure devel-
oped through it. Similar pattern was observed in FLAC3D
model. However, it was found from the model that the
crocsscut becomes unstable after excavating the third cut
Figure 18. Yield pattern for a non-linear
FLAC3D models of crosscut development using
both the first and the second methods. Yielded
elements are shown in red
not the fourth cut as found from field observation. Field
observations showed that when the first method was used,
the crosscut remained stable during the first three cuts.
However, instability and cutter failure developed after the
final cut. The FLAC3D model displayed a similar pattern,
although it indicated instability occurring after the third
cut rather than the fourth, as observed in the field. Note
that the direction of the maximum horizontal stress is par-
allel to the Y-axis.
SUMMARY AND CONCLUSION
Field observations from an operating mine and numerical
models were used to assess the impact of maximum hori-
zontal stress orientation, caprock thickness, face advance
methods, and cutting sequences on roof stability. The key
findings are as follows: when the caprock is thin and the
roof is subjected to high horizontal stress, it is highly rec-
ommended to align entries parallel to the maximum hori-
zontal stress to mitigate the detrimental effect of excessive
horizontal stress. However, if the caprock is thick and mas-
sive, the influence of maximum horizontal stress on roof
stability is less significant, even in unfavorable orientations,
as increased caprock thickness reduces the induced shear
stress in the roof above the excavation. Both field observa-
tions and numerical models suggest avoiding mining with
an arrowhead front, as this method causes significant dam-
age to all entries. In contrast, with a flat-front advance, only
the outer entries were damaged. These findings enhance
understanding of roof stability under high horizontal stress
and contribute to reducing the risk of roof falls in under-
ground stone mines. Additionally, preliminary evidence
suggests enhanced stability when creating crosscuts by con-
necting existing entries to minimize instabilities caused by
high horizontal stress.
LIMITATIONS OF THE STUDY
Many of the hazardous conditions in underground stone
mines are the result of a complex interaction between geo-
logic conditions and mining-induced factors. The conclu-
sions drawn from the field observations in this study are
inherently limited by the specific mining and geological
conditions present at the study site, which may or may not
be directly applicable to other mines with different charac-
teristics. Variations in rock mass properties, stress regimes,
and geological structures in other mines could lead to dif-
ferent outcomes.
Additionally, the results obtained from the FLAC3D
models are constrained by the input parameters and the
applied boundary conditions. While every effort was made
to use realistic and representative values, uncertainties in
creating a crosscut using the first method led to instability,
while the second method produced a more stable crosscut.
As illustrated in Figure 18, the yield pattern in the immedi-
ate roof from the FLAC3D model for the second method is
significantly less than that of the first method. Therefore, it
is recommended to develop new crosscuts using the second
method when the roof is subjected to high horizontal stress.
Based on field observation, when the first method was
used to create a crosscut, the crosscut was stable during
the first three cuts however, when the last cut was done,
the crosscut becomes unstable and cutter failure devel-
oped through it. Similar pattern was observed in FLAC3D
model. However, it was found from the model that the
crocsscut becomes unstable after excavating the third cut
Figure 18. Yield pattern for a non-linear
FLAC3D models of crosscut development using
both the first and the second methods. Yielded
elements are shown in red
not the fourth cut as found from field observation. Field
observations showed that when the first method was used,
the crosscut remained stable during the first three cuts.
However, instability and cutter failure developed after the
final cut. The FLAC3D model displayed a similar pattern,
although it indicated instability occurring after the third
cut rather than the fourth, as observed in the field. Note
that the direction of the maximum horizontal stress is par-
allel to the Y-axis.
SUMMARY AND CONCLUSION
Field observations from an operating mine and numerical
models were used to assess the impact of maximum hori-
zontal stress orientation, caprock thickness, face advance
methods, and cutting sequences on roof stability. The key
findings are as follows: when the caprock is thin and the
roof is subjected to high horizontal stress, it is highly rec-
ommended to align entries parallel to the maximum hori-
zontal stress to mitigate the detrimental effect of excessive
horizontal stress. However, if the caprock is thick and mas-
sive, the influence of maximum horizontal stress on roof
stability is less significant, even in unfavorable orientations,
as increased caprock thickness reduces the induced shear
stress in the roof above the excavation. Both field observa-
tions and numerical models suggest avoiding mining with
an arrowhead front, as this method causes significant dam-
age to all entries. In contrast, with a flat-front advance, only
the outer entries were damaged. These findings enhance
understanding of roof stability under high horizontal stress
and contribute to reducing the risk of roof falls in under-
ground stone mines. Additionally, preliminary evidence
suggests enhanced stability when creating crosscuts by con-
necting existing entries to minimize instabilities caused by
high horizontal stress.
LIMITATIONS OF THE STUDY
Many of the hazardous conditions in underground stone
mines are the result of a complex interaction between geo-
logic conditions and mining-induced factors. The conclu-
sions drawn from the field observations in this study are
inherently limited by the specific mining and geological
conditions present at the study site, which may or may not
be directly applicable to other mines with different charac-
teristics. Variations in rock mass properties, stress regimes,
and geological structures in other mines could lead to dif-
ferent outcomes.
Additionally, the results obtained from the FLAC3D
models are constrained by the input parameters and the
applied boundary conditions. While every effort was made
to use realistic and representative values, uncertainties in