Page 80 - Advances In Productive, Safe, and Responsible Coal Mining

66 Advances in Productive, Safe, and Responsible Coal Mining

Once the model is constructed, the next important step is to ensure that it behaves as
intended and accurately predicts the defined output. Model verification is usually done
using animation to evaluate the behavior of the system. To validate the model, the
simulation output (production rate) should be compared with real mine data to ensure
that model predictions are within acceptable limits. The model can be used for further
experimental analysis once validated.

5.2.1.2 Determine feasible range of input variables

The analyst determines the set of feasible scenarios by determining possible values of
input variables given the existing mining constraints. A full factorial experimental
design approach is then used to evaluate each combination of input variables in the
feasible set. To understand the relationship between panel width and production rate,
the primary input variable is panel width. However, the analyst must also specify the
cut sequence and equipment fleet for each panel width since these input variables also
affect the production rate.
The total number of experiments (scenarios) depends on system specifications,
constraints, and stakeholders’ expectations. For example, given the fleet size for an
existing mine, if the analyst was to include relatively small panel widths, these will
lead to long queues at the CM, which are likely to lead to exceedingly large cycle
times. This makes it impractical to include relatively small panel widths in the feasible
set as production rates are too low to be considered practical.

5.2.1.3 Estimate production rates
In this step, the analyst conducts simulation experiments for each of the scenarios
identified in the previous step. The number of replications required for each scenario
depends on the uncertainty associated with predictions of production rate in the par-
ticular instance. Often, the number of replications is determined based on the half-
width (an estimate of the confidence interval assuming a normal distribution) [10].
Production rates can then be used in the optimization of recovery and production rates
that is discussed in Section 5.3. First, a case study is described to illustrate the
approach described in this section.

5.2.2 Case study
The case study used in this chapter is an underground R&P coal mine located in south-
ern Illinois, the United States. The mine produces approximately 7 million tons of coal
at a 54% panel recovery rate. The mine has experimented with multiple panel widths
ranging from 11 to 21 entries. Each panel is a supersection mined with two CMs. Joy
Model 14CM27 continuous miners are used in each section along with four 20ton Joy
Model BH20 battery-powered haulage units that transport coal from the cutting face to
the FB. The CM cuts and loads coal at up to 40ton/min with a maximum cutting height
of 11.2ft. Full panel width is mined in six-crosscut increments with the FB moved in
three-crosscut increments. The FB is located at the center of each production panel to