Top Banner
ORIGINAL PAPER Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine— A Case Study Piyush Rai Umakant Yadav Ashok Kumar Received: 6 April 2009 / Accepted: 21 March 2011 / Published online: 3 May 2011 Ó Springer Science+Business Media B.V. 2011 Abstract The tandem operation of draglines is in use in some of the major Indian opencast coalmines owing to the favourable geo-mining conditions, technical suitability, efficiency and economic viabil- ity. In view of the importance of tandem operation, the present study has been undertaken in a large Indian opencast coalmine in order to critically investigate the horizontal and vertical tandem oper- ation of draglines on moderately strong and high sandstone benches (35–42 m), overlying a 15–18 m thick coal seam. The study has revealed that although the preparation of balancing diagram for planning of dragline operations is the first and the most important step, its actual implementation is equally important. Marked discrepancies in the productivity parameters as envisaged by the balancing diagram and as observed in the field studies, have been investigated and reported. The study also propounds the impor- tance of appraisal of dragline productivity parame- ters, such as, cycle time, swing angle, seating position, availability, utilization, etc., in the field scale. Irrespective of the mode of operation (hori- zontal or vertical tandem), the study moots the concept of computation of the weighted cycle time and overall cycle time vis-a `-vis swing angle variation for the draglines operating in field. The results drawn from the case study have been discussed in terms of cycle time computations, annual output computation and evaluation of earthmoving efficiency for the horizontal and vertical tandem modes of operation. Keywords Tandem operation Dragline balancing diagram Weighted cycle time Swing angle Loose muck Blasted muck 1 Introduction The walking draglines offer several merits in open pit mining project where long reach, deep digging and high output are essential requisites and the volume of overburden (O/B) to be handled is many times in comparison to the volume of mineable mineral. For instance, a 1 m thick coal seam may have 30 m thick O/B cover which may still be an economic P. Rai (&) Department of Mining Engineering, Institute of Technology, Banaras Hindu University, Varanasi 221 005, India e-mail: [email protected]; [email protected] P. Rai Department of Energy and Resources Engineering, Chonnam National University, Gwangju, South Korea U. Yadav Corporate Office, Northern Coalfields Ltd. (a subsidiary of Coal India Ltd.), Singrauli, Dist. Sidhi, Madhya Pradesh, India A. Kumar Pakri Barawdih Project, National Thermal Power Corporation, Dist. Ranchi, Jharkhand, India 123 Geotech Geol Eng (2011) 29:493–504 DOI 10.1007/s10706-011-9398-9
12

Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

Jan 22, 2023

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

ORIGINAL PAPER

Productivity Analysis of Draglines Operating in Horizontaland Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

Piyush Rai • Umakant Yadav • Ashok Kumar

Received: 6 April 2009 / Accepted: 21 March 2011 / Published online: 3 May 2011

� Springer Science+Business Media B.V. 2011

Abstract The tandem operation of draglines is in

use in some of the major Indian opencast coalmines

owing to the favourable geo-mining conditions,

technical suitability, efficiency and economic viabil-

ity. In view of the importance of tandem operation,

the present study has been undertaken in a large

Indian opencast coalmine in order to critically

investigate the horizontal and vertical tandem oper-

ation of draglines on moderately strong and high

sandstone benches (35–42 m), overlying a 15–18 m

thick coal seam. The study has revealed that although

the preparation of balancing diagram for planning of

dragline operations is the first and the most important

step, its actual implementation is equally important.

Marked discrepancies in the productivity parameters

as envisaged by the balancing diagram and as

observed in the field studies, have been investigated

and reported. The study also propounds the impor-

tance of appraisal of dragline productivity parame-

ters, such as, cycle time, swing angle, seating

position, availability, utilization, etc., in the field

scale. Irrespective of the mode of operation (hori-

zontal or vertical tandem), the study moots the

concept of computation of the weighted cycle time

and overall cycle time vis-a-vis swing angle variation

for the draglines operating in field. The results drawn

from the case study have been discussed in terms of

cycle time computations, annual output computation

and evaluation of earthmoving efficiency for the

horizontal and vertical tandem modes of operation.

Keywords Tandem operation � Dragline balancing

diagram �Weighted cycle time � Swing angle � Loose

muck � Blasted muck

1 Introduction

The walking draglines offer several merits in open pit

mining project where long reach, deep digging and

high output are essential requisites and the volume of

overburden (O/B) to be handled is many times in

comparison to the volume of mineable mineral. For

instance, a 1 m thick coal seam may have 30 m thick

O/B cover which may still be an economic

P. Rai (&)

Department of Mining Engineering, Institute

of Technology, Banaras Hindu University,

Varanasi 221 005, India

e-mail: [email protected];

[email protected]

P. Rai

Department of Energy and Resources Engineering,

Chonnam National University, Gwangju, South Korea

U. Yadav

Corporate Office, Northern Coalfields Ltd.

(a subsidiary of Coal India Ltd.), Singrauli, Dist.

Sidhi, Madhya Pradesh, India

A. Kumar

Pakri Barawdih Project, National Thermal Power

Corporation, Dist. Ranchi, Jharkhand, India

123

Geotech Geol Eng (2011) 29:493–504

DOI 10.1007/s10706-011-9398-9

Page 2: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

proposition with application of draglines. Further,

absence of auxiliary haulage units, higher mineral

recovery, easy maneuverability, low maintenance

costs and operational advantages under adverse pit

conditions are also cited as the merits of deploying

draglines (Chugh 1980).

In the past two decades, draglines have especially

gained vast popularity in India for O/B stripping in

opencast coalmines. At present, Indian coalmines

have about 43 draglines in operation, which handle

almost 22% of the O/B. The bucket sizes of these

draglines range from 5 to 30 m3 (Balamadeswaran

et al. 2004). However, it must be clearly understood

that dragline is the single most expensive piece of

equipment to be deployed by any opencast mine,

hence, it is imperative to keep a strict vigil on the

performance of this equipment under the prevalent

operating conditions.

Considering the importance of the dragline oper-

ations and the costs involved therein, the present

work aims at critically investigating the dragline

performance in two different modes of tandem

operation, namely horizontal tandem (HT) and ver-

tical tandem (VT) operation (on same O/B strata and

under similar conditions). The study is based on

rigorous field monitoring in order to quantitatively

assess the key productivity parameters, while plan-

ning the tandem operations.

1.1 Objectives of the Study

The specific objectives of the present case study are

enumerated as follows:

1. To construct the balancing diagram for HT as

well as VT modes of operation under given field

conditions and to subsequently compute the

annual output by using the constructed balancing

diagram.

2. To compute some important productivity param-

eters, through field studies, for cut widths of 90

and 85 m, as practiced in field for HT and VT

modes of operation, respectively.

3. To study and establish the relationship between

varying swing components and the overall cycle

time of the draglines operating in tandem (HT as

well as VT).

4. To investigate the influence of overall cycle time,

availability, utilization and cut width on the

performance parameters of draglines in the HT

and VT modes of operation.

5. To assess the possible reasons for variation in

the productivity parameters as obtained from the

constructed balancing diagram, vis-a-vis the

actual operation in field.

2 Case Description

The present case study has been carried out in mine A

of a major opencast coalmine of Northern Coalfields

Ltd., Singrauli, India. The mine A is divided by a

central entry into two operating sections along the

East–West strike of the property. In one of the sections,

which will henceforth be named as mine 1, draglines

were operating in the HT mode of operation, whereas,

in another section, which will henceforth be named as

mine 2, draglines were operating in VT mode of

operation. Draglines operated on drilled and blasted

high O/B benches, which consisted of medium to fine

grained sandstone. The thickness of underlying coal

seam ranged from 15 to 18 m. The draglines casted the

overburden into previously cut void created after

extraction of the underlying coal seam.

2.1 Description of Horizontal Tandem Operation

in Mine 1

The cross-section and plan view of a typical horizontal

tandem operation is illustrated in Figs. 1 and 2

respectively. As clearly illustrated, both 24/96 drag-

lines were deployed on the same stripping bench (35 m

high) to work in horizontal tandem. The leading

dragline (dragline no. 1, LeD/L-HT) was deployed on

the highball side to provide the key cut towards the

highball. On completing the key cut this dragline

excavated the first cut (next to key cut). After

excavating these two cuts, the leading dragline again

moved to new key cut position, ready for next stripping

cycle to repeat the sequence. The lagging dragline

(dragline no. 2, LaD/L-HT), staggered at least 200 m

behind the leading dragline (due to safety reasons), sat

on the spoil side of the extended bench formed by

leading dragline. Being seated on the extended bench it

side casted the O/B to a greater distance. This dragline

excavated the first dig (next to first cut) on the

remaining portion of the stripping bench, and also

494 Geotech Geol Eng (2011) 29:493–504

123

Page 3: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

re-handled the loose overburden to finally expose the

coal seam fully as per the designed balancing diagram

parameters. Both draglines advanced along the strike

away from the central entry towards the boundary of

the pit to make cuts. A coal rib, of 5 m width

(triangular section) at the bottom, and up to full seam

thickness was left against the spoil heap.

2.2 Description of Vertical Tandem Operation

in Mine 2

In order to remove very thick overburden cover (in the

form of 42 m high bench), overlying the coal seam, the

VT mode of operation was practiced in the mine 2.

The cross-section and plan view of a typical vertical

tandem operation is illustrated in Figs. 3 and 4,

respectively. In this mode of operation, the 42 m high

bench was divided into two vertical benches, viz,

upper bench (top bench) and lower bench (main

bench). The upper bench height was kept 14 m high

and the lower bench height was 28 m. A 15/90

dragline (dragline 1, LeD/L-VT) seated on upper

bench excavated the O/B and side casted the spoil into

the de-coaled area which extended the main bench.

After sufficient advance of the top dig the same

dragline came down to main bench for excavating the

key cut and side casting the spoil into the de-coaled

area to form a level pad to create seating space for

lagging dragline after which, it again moved to the top

bench for next striping cycle. The lagging 24/96

dragline (dragline 2, LaD/L-VT) on the lower bench

excavated the first dig portion and subsequently

Fig. 1 Cross-section

showing draglines in

horizontal tandem

Fig. 2 Plan view of

draglines operating in

horizontal tandem

Geotech Geol Eng (2011) 29:493–504 495

123

Page 4: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

marched towards the spoil side to sit on the level pad

(extended bench) to re-handle the spoil finally and

fully exposes the coal seam

3 Study Methodology

In this study, balancing diagrams for the HT and VT

modes of operation was constructed, to the scale, by

using the key field parameters (as tabulated in

Tables 1 and 2). The constructed balancing diagram

was used to estimate the annual output by the

draglines in HT and VT modes of operation. There-

after, a rigorous field study was undertaken to collect

the operational cycle time, sitting position, availabil-

ity and utilization factors for a substantially long

period of time in order that the collected data could

be used as representative data for projection of annual

Fig. 3 Cross-section

showing draglines in

vertical tandem

Fig. 4 Plan view of draglines operating in vertical tandem

496 Geotech Geol Eng (2011) 29:493–504

123

Page 5: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

output, and, consequently for computing the earth

moving efficiencies in HT and VT modes of

operation.

4 Construction of Balancing Diagram

Based on the standard procedures as suggested by Lal

et al. (1988), Rai (1997) and CMPDIL Report (1998),

the balancing diagram for draglines operating in HT

and VT modes of operation was constructed to the

scale, by using the given key field parameters, as

tabulated in Tables 1 and 2, respectively. The con-

structed balancing diagrams are illustrated in Figs. 5

and 6 for HT and VT modes of operation, respectively.

These constructed balancing diagrams were used

as a tool for computation of the desired annual O/B

output, annual coal exposure, linear advance and the

re-handling percentage for the respective modes of

operation.

5 Field Observations

Present case study was undertaken on pilot scale in

mine 1 for HT mode and mine 2 for VT mode of

dragline operation to gather a sound representative

field data for ensuing investigations. Field data

collection procedure is enumerated and discussed

step-by-step as follows.

Table 1 Key field

parameters and machine

parameters for construction

of balancing diagram in

horizontal tandem (HT)

mode of operation in mine 1

Sl. No. Parameters Details

1 Draglines

D/L-1(LeDL-HT) 24/96 with max. horizontal reach of 88 m

D/L-2(LaDL-HT) 24/96 with max. horizontal reach of 88 m

2 Bench height Approx. 35 m

3 Cut width 90 m

4 Strip length 1,700 m

5 Highwall slope 70�6 Bench slope 60�7 Key cut width (at top) 38 m

8 Key cut width (at bottom) 5 m

9 Coal rib Left for full seam height (triangular

section)

10 Angle of repose 38�

Table 2 Key field

parameters and machine

parameters for construction

of balancing diagram in

vertical tandems (VT) mode

of operation in mine 2

Sl. No. Parameters Details

1 Draglines

D/L-1(LeDL-VT) 15/90 with max. horizontal reach of 82 m (top portion)

D/L-2(LaDL-VT) 24/96 with max. horizontal reach of 88 m (bottom portion)

2 Bench height Approx. 42 m (14 m as top portion and 28 m as bottom

portion)

3 Cut width 85 m

4 Strip length 2,000 m

5 Highwall slope 70�6 Bench slope 60�7 Key cut width at top 38 m

8 Key cut width at bottom 5 m

9 Coal rib Left for full height (triangular section)

10 Angle of repose 38�

Geotech Geol Eng (2011) 29:493–504 497

123

Page 6: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

5.1 Collection of Data for Cycle Time,

Availability and Utilization Results

A typical dragline operation cycle can be distinctly

splitted into four discrete segments namely, digging

segment, swing forward segment, unloading seg-

ment, and swing back and bucket re-positioning

segment. Stopwatch was used to record the cycle

times for all the four draglines. Actual work hours

(WH), idle hours (IH), maintenance hours (MH)

and breakdown hours (BH) were registered, for all

the draglines during the study period, to evaluate

the representative data for availability (A) and

utilization (U). These representative field data on

cycle time, A and U were used for projecting the

annual output of the draglines.

5.2 Computations of Weighted and Overall

Cycle Time

The well established method of time and motion

study was adopted for data collection. During the

course of collection of cycle time data for both the

draglines operating in HT mode of operation, it

was observed that being seated at one position, the

draglines operated at swing angles varying between

90� and 180� to cast the muck (blasted overbur-

den). The percentages of the muck removed, from

one sitting position by the draglines, were almost

50% at swing angle of 90�, 25% in between 90�and 120� swing, 15% in between 120� and 150�swing and 10% in between 150� and 180� swing.

Almost similar trends were observed on both the

draglines operating in the VT mode of operation,

wherein, it was observed that being seated at one

position, the draglines again operated at swing angles

varying between 90� and 180�. The percentages of

the blasted muck removed from one sitting position

were almost 50% at swing angle up to 90�, 26% in

between 90� and 120�, 14% in between 120� and

150� and 10% in between 150� and 180�.

Looking at the percentages of material removed by

the draglines, irrespective of the operational modes,

Eq. 1 was developed by the authors to compute the

Fig. 5 Balancing diagram

and related computations

for draglines operating in

vertical tandem

498 Geotech Geol Eng (2011) 29:493–504

123

Page 7: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

weighted cycle time. Nevertheless, Eq. 1 is case

specific, depending on various factors such as drag-

line sitting positions, design parameters for balancing

diagram, rock characteristics, blasting efficiency and

the operator’s skill. As such, it may be consequential

to mention here that religious time and motion study

need to be carried for generation of such equations on

case-to-case basis.

Weighted cycle time CWð Þ¼ 50�C1 þ 25�C2 þ 15�C3 þ 10�C4ð Þ=100 ð1Þ

where C1 is the average total cycle time up to 90�swing, C2 the average total cycle time for 90–120�swing, C3 the average total cycle time for 120–150�swing, and C4 is the average total cycle time for

150–180� swing. A generalized equation can be put

into simple form as:

Weighted cycle time CWð Þ¼ k�1C1 þ k�2C2 þ k�3C3 þ k�4C4=100� �

ð1aÞ

where k1, k2, k3 and k4 are observed percentages of

occurrence of average total cycle time for a specified

range of swing angle such that k1 ? k2 ? k3 ?

k4 = 100.

The overall cycle time was then computed for each

dragline on the basis of computed weighted cycle

time for draglines operating in blasted muck and in

re-handled muck (loose O/B).

The re-handling operation, as also evident from the

dragline balancing diagrams (Figs. 5, 6), is done only

by the lagging draglines. During the field studies it

was estimated that approximately 40% of the total

muck handling time was spent on re-handling the

loose O/B (muck). The lagging dragline spent almost

60% of the total muck handling time on handling the

blasted muck. Since the total cycle time for excavat-

ing the loose O/B (re-handling) was always less than

the total cycle time for excavating the blasted muck,

the concept of overall cycle time (COA) for lagging

draglines has been quantified to take care of

Fig. 6 Balancing diagram

and related computations

for draglines operating in

horizontal tandem

Geotech Geol Eng (2011) 29:493–504 499

123

Page 8: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

fractional cycle time in the loose and the blasted

muck portions. The COA as quantified by Eq. 2, is

case specific.

Overall cycle time COAð Þ ¼ 0:4�CWLð Þ þ 0:6�CWBð Þð2Þ

where CWL is the weighted cycle time for re-handled

muck (loose O/B) and CWB is the weighted cycle

time for blasted muck. A generalized equation can be

put into simple form as:

Overall cycle time COAð Þ ¼ kCWL þ 1� kð ÞCWB

ð2bÞ

where k is fraction of time spent for re-handling the

loose muck by lagging draglines. Further, since the

leading dragline always operated on the blasted muck

only, the overall cycle time for leading draglines COA

will always equal to CW.

5.3 Evaluation of Availability and Utilization

After collection of field data on WH, BH and IH, the

availability, utilization and availability-cum-utiliza-

tion factor, for all the draglines under study, were

evaluated by using the standard equations given

below:

Availability factor Að Þ ¼ SSH � MHþ BHð ÞSSH

ð3Þ

Utilization factor Uð Þ ¼ SSH � MHþ BHþ IHð ÞSSH

ð4ÞAvailability - cum - utilization factor Kð Þ ¼ A� U

ð5Þ

where SSH is the scheduled shift hour (in a specified

period, where 1 shift = 8 h), MH the maintenance

hour, IH the idle hour and BH is the breakdown hour.

5.4 Projection of Annual Output

On the basis of the collected field data, the projected

annual output (P1) for each dragline was computed by

using the standard Eq. 6 as given below (CMPDIL

norms 1998).

P1 ¼ B=Cð Þ�K�S�F�M�N�h N�s N�d 60�60 M m3� �

ð6Þ

where B is the bucket capacity of dragline (m3), C the

overall cycle time of dragline (s), K the availability-

cum-utilization factor, S the swell factor for loose, easy

digging sandstone (0.719), F the fill factor for loose,

easy digging sandstone (0.933), M the machine travel-

ling and positioning factor (0.8), Nh the number of hours

in a shift (8 h), Ns the number of shifts in a day (3 shifts),

and Nd is the number of days in a year (365 days).

In order to compute the annual outputs for the

draglines under study, the individual values of overall

cycle time (COA), A and U were substituted for each

dragline as per the recorded field observations. Remain-

ing factors in Eq. 6 were substituted as per the Central

Mine Planning & Design Institute Ltd. (CMPDIL)

norms as given above within the parentheses.

5.5 Computation of Efficiency for Draglines

The computation of earthmoving efficiency for each

dragline was done by using Eq. 7, as suggested by

Rai (1997).

Efficiency of dragline gð Þ

¼ Computed annual output P1ð Þ � 100

Annual output as per the balancing diagram ðP2Þð7Þ

The determination of annual output as per the balanc-

ing diagram (P2) was done from the prepared balancing

diagram as per the balancing results given in Figs. 5

and 6 for HT and VT modes of operation, respectively.

5.6 Computation of Coal Exposure

The coal exposed by the draglines working in tandem

operation was estimated by using the generalized

Eq. 8 as per the balancing diagram concept.

CE ¼ PFD=Að Þ �W � T � D� R M teð Þ ð8Þ

where CE is the coal exposure (M te), PFD the annual

output of the lagging dragline from the first dig (M

m3), A the area of first dig (m2), W the cut width (m),

T the thickness of coal seam (m), D the specific

gravity of coal, and R is the recovery factor.

In Eq. 8 the term (PFD/A) represents the annual

linear advance of the draglines (more specifically

lagging draglines). In the above equation, the value of

W was fixed as per the mode of operation (90 and

85 m for HT and VT operation, respectively) and the

500 Geotech Geol Eng (2011) 29:493–504

123

Page 9: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

value of T was taken as 18 m for HT operation and

15 m for VT mode. D was taken as 1.52 for the given

field conditions, whereas, the value of R was assumed

as 0.9. Thereby, on substituting the values of the

linear advance (PFD/A), as obtained from the balanc-

ing diagram the coal exposure value was obtained

(refer to balancing results as given in Figs. 5 and 6).

6 Results and Discussion

6.1 Cycle Time Results

Overall cycle time (COA) for individual draglines was

computed on the basis of weighted values at different

swing angles for blasted muck and re-handled muck

(loose O/B), by using Eqs. 1 and 2, respectively. The

cycle time results are tabulated in Tables 3, 4, 5, 6.

Cycle time results, as tabulated in Tables 3, 4, 5, 6,

reveal that, irrespective of the operational mode of

draglines, weighted cycle time on re-handled muck is

less than weighted cycle time on blasted muck. This

is due to the easier diggability in the re-handled muck

owing to its looseness. Hence, it implies that a greater

volume of muck can be handled by lagging draglines

as besides operating on the blasted muck in the first

dig area, they operate on the loose muck also.

Further, overall cycle time results, as seen in above

tables, also clearly indicate that the leading draglines

have higher overall cycle time in comparison to

Table 3 Weighted and overall cycle time results for 24/96 leading dragline (LeHT-D/L) working in HT mode in the blasted muck of

mine 1

Sl. No. Mode of

operation

Swing angle (�) Observed cycle time

for 24/96 (LeHT-D/L) (s)

Weighted cycle

time (CW) (s)

Overall cycle time (COA)

for 24/96 (LeHT-D/L) (s)a

1 HT Up to 90 74.58 81.48 81.48

2 HT [90–120 81.80

3 HT [120–150 91.89

4 HT [150–180 99.53

a In the HT as well as the VT modes of operation since the leading draglines (Le D/L) operate on the blasted muck hence, CW = COA

(since k = 0)

Table 4 Weighted and overall cycle time for 24/96 lagging dragline (LaHT-D/L) working in HT mode in the blasted muck and re-

handled muck (loose O/B) for mine 1

Sl. No. Mode of

operation

Swing

angle (�)

Observed cycle

time for 24/96

(LaHT-D/L) for

blasted muck (s)

Observed cycle

time for 24/96

(LaHTD/L) for

re-handled muck (s)

Weighted

cycle time

for blasted

muck (s)

Weighted

cycle time

for re-handled

muck (s)

Overall cycle

time (COA)

for 24/96

(LaHT-D/L) (s)

1 HT Up to 90 74.96 65.11 81.16 71.25 77.20

2 HT [90–120 81.68 73.93

3 HT [120–150 89.45 78.57

4 HT [150–180 98.40 84.29

Table 5 Weighted and overall cycle time for 15/90 leading dragline (LeVT-D/L) working in VT mode in the blasted muck of mine 2

Sl. No. Mode of

operation

Swing angle (�) Observed cycle time

for 15/90 (LeVT-D/L) (s)

Weighted cycle

time (CW) (s)

Overall cycle time (COA)

for 15/90 (LeVT-D/L) (s)a

1 VT Up to 90 74.58 80.58 80.58

2 VT [90–120 81.03

3 VT [120–150 89.38

4 VT [150–180 96.26

a In the HT as well as the VT modes of operation since the leading draglines (Le D/L) operate on the blasted muck hence, CW = COA

(since k = 0)

Geotech Geol Eng (2011) 29:493–504 501

123

Page 10: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

lagging draglines. This is again attributable to the fact

that leading draglines always operate only on the

blasted muck which is always harder to dig in

comparison to the re-handled muck. The looseness

factor enables the handling of muck in less time.

Though swell factor is being taken into consideration

for conversion of removed material in terms of solid

volume of muck and all the material handled is

expressed in terms of solid volume, the increased

operational efficiency achieved due to decrease in

weighted cycle time in loose muck re-handling needs

due consideration. It may be essential to consider this

feature while planning the tandem balancing opera-

tions, which only considers the ratio of bucket

capacities for the workload distribution on respective

draglines irrespective of their seating locations (on

blasted or loose O/B).

Another important feature, which is noteworthy

from the cycle time results, is the increased weighted

and overall cycle times, for the draglines under study,

in comparison to the standard norm of 60 s, which is

used for estimating the dragline output by the

balancing diagram concept. During the field moni-

toring, it was observed that from one seating position,

the draglines in the HT as well as VT modes operated

at swing angles varying from 90� to 180�. Although

swing angles up to 90� is standard norm, the variation

up to 120� has been cited, and it has been reported

that increase in swing angle obviously lead to

increase in swing time. Earlier research works by

Gupta (1981), Mathur (1999), Naganna and Rai

(2003), Rai (2004) and Kishore (2004) reported the

swing angle variation from 90� to 180� in the

opencast mines. Higher swing angle inordinately

increases the cycle time of the draglines, which is

counter productive to its productivity. During the

field studies, it was experienced that in order to

handle maximum amount of blasted O/B from one

seating location, the dragline marching was compro-

mised with the increase in the swing angle and swing

time, which, in turn, increased the cycle time

exorbitantly. To this end, a judicious approach is

necessitated to strike a balance between seating

location and loss of time in marching and re-

positioning of dragline, after considering the site-

specific constraints.

6.2 Availability and Utilization Results

The field observations pertaining to the break up of

scheduled shift hours were analyzed, and the com-

putations for actual A, U and K were done by using

Eqs. 3, 4 and 5. The results for A, U and K are

presented in Tables 7 and 8.

The results of Tables 7 and 8 indicate that the

values of A and U are low for all the draglines in

comparison to the desirable standards. It is conse-

quential to mention at this stage that, the draglines

have been reported to perform at availability as well

as utilization levels of 0.90–0.95 (Chironis 1986).

Table 6 Weighted and overall cycle time for 24/96 lagging dragline (LaVT-D/L) in the blasted muck and re-handled muck (Loose

O/B) of mine 2

Sl. No. Mode of

operation

Swing

angle (�)

Observed cycle

time for 24/96

(LaVT-D/L) for

blasted muck (s)

Observed cycle

time for 24/96

(LaVT-D/L) for

re-handled muck (s)

Weighted

cycle time

for blasted

muck (s)

Weighted

cycle time

for re-handled

muck (s)

Overall cycle

time (COA)

for 24/96

(LaVTD/L) (s)

1 VT Up to 90 76.67 65.11 83.06 70.70 78.12

2 VT [90–120 83.26 73.07

3 VT [120–150 94.97 76.27

4 VT [150–180 96.69 84.35

Table 7 Computed availability, utilization and availability-cum-utilization factor for draglines in mine 1

Equipment Mode of

operation

Availability

factor (A)

Utilization

factor (U)

Availability-cum-utilization

factor (K) under field study

24/96 (LeHT-D/L) HT 0.8883 0.8288 0.7318

24/96 (LaHT-D/L) HT 0.8913 0.8134 0.7249

502 Geotech Geol Eng (2011) 29:493–504

123

Page 11: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

Further, it is also evident from the above tables that

the significant discrepancy between the A and U levels

is due to unnecessary idling and large swing angles of

the draglines in the field. The down time, delays, and

thus the performance statistics need careful planning

for increasing dragline productivity.

6.3 Results of Annual Output and Earthmoving

Efficiencies

On the basis of designed balancing diagram, field-

work and related computations using Eqs. 6 and 7,

results of annual output and the earthmoving effi-

ciencies for the draglines are tabulated in Table 9.

From the results given in Table 9, the computed

annual output on the basis of field observations for

HT operation is 6.92 M m3 whereas output as per the

balancing diagram comes out to 7.94 M m3 the ratio

of these two outputs provides the earthmoving

efficiency for the dragline in HT mode, which is

87.15%. Similarly, from the results of Table 9, the

computed annual output on the basis of field obser-

vations for VT operation is 5.37 M m3 whereas

output as per the balancing diagram comes out to

6.26 M m3, hence, the earthmoving efficiency is

85.78% in the VT mode. From the output and

efficiency results, it is quite evident that irrespective

of mode of operation, there is substantial gap between

the actual performances of the dragline in the field

against its planned performance as per the balancing

diagram.

The aforementioned results and discussions call

for serious thinking and suitable modification in

planning the production and productivity issues for

draglines operating in tandem. Such studies are the

need of the hour to improve upon the dragline

productivity. Although it is beyond the scope of the

present paper, it may be of consequence to mention

here that in order to improve the productivity levels

and patch up the prevalent discrepancies, a three-

dimensional balancing diagram may be contemplated

to be more precise and pragmatic under the given

field conditions. The futuristic researches may con-

sider this viewpoint.

7 Conclusions

The study presents following conclusions:

Balancing diagram is a very important tool to plan

and analyze the key performance parameters for

draglines. It must be used as a useful tool for

evaluation of dragline performance. The discrepan-

cies between the balancing diagram results and the

results obtained from actual field performance study

must be thoroughly examined in order to improve the

productivity. In the present case, through the rigorous

field monitoring the discrepancies were examined in

swing angle, overall cycle time, availability, utiliza-

tion and differential overall cycle time in blasted and

re-handled mucks.

The study focuses on the concept of assigning due

weightings to varying swing component in order to

develop the governing equation for computation of

overall cycle time. The governing equations for the

computation of overall cycle time are very likely to

Table 8 Computed availability, utilization and availability-cum-utilization factor for draglines in mine 2

Equipment Mode of

operation

Availability

factor (A)

Utilization

factor (U)

Availability-cum-utilization

factor (K) under field study

15/90 (Le VT-D/L) VT 0.877 0.81 0.7104

24/96 (La VT-D/L) VT 0.85 0.7916 0.6730

Table 9 Computed annual output vis-a-vis annual output proposed by the designed balancing diagram (HT and VT mode)

Mine Mode of

operation

Output as per balancing

diagram (M m3)

Computed output from the

field observations (M m3) using Eq. 6

Earthmoving efficiency

(P1/P2) 9 100 (%)

Total (P2) Total (P1)

Mine 1 HT 7.94 6.92 87.15

Mine 2 VT 6.26 5.37 85.78

Geotech Geol Eng (2011) 29:493–504 503

123

Page 12: Productivity Analysis of Draglines Operating in Horizontal and Vertical Tandem Mode of Operation in a Coal Mine—A Case Study

be case specific as they depend on the geo-mining

conditions, degree of fragmentation, machine related

parameters, and, above all the management decisions.

However, the critical investigations of cycle time vis-

a-vis swing variation would be of immense value to

the field planners and operators to clearly investigate

and optimize the cycle time for enhanced

productivity.

Looking at the differential overall cycle times for

the leading and lagging draglines, it may be reason-

able to give due cognizance to overall cycle time

while working on the blasted muck, and, while

working on the re-handling portion of the blasted

muck (which is quite loose and easier to handle).

The swing angle is one of the major contributors

towards the enhanced overall cycle time, hence

efforts must be made to minimize the swing angle

instead of minimizing the marching of draglines.

Unnecessarily trying to reduce the dragline marching

results into excessive swing angle and swinging time,

which unnecessarily escalates the overall cycle time.

Availability and utilization of the draglines must

be properly recorded, documented and monitored on

a regular basis in order to facilitate continual

improvement in the productivity of capital-intensive

draglines.

Acknowledgments The author wishes to convey immense

gratitude towards the staff and management of the Northern

Coalfields Ltd. (NCL), Singrauli, Dist. (MP), India, for

providing their permission and excellent support during the

fieldwork.

References

Balamadeswaran P, Sen P, Mishra AK (2004) Blasting prac-

tices in Indian dragline benches: a review. Minetech J

CMPDIL 25(4):36–44

Chironis NP (1986) Smaller versatile draglines operating in

tandem reach deep seams efficiently. Coal Age 9(12):

48–50

Chugh YP (1980) Surface mining of minerals by draglines in

the USA. Transaction of mining metallurgy (Sectional

Mining Industry). The Institution of Mining and Metal-

lurgy, London, pp A198–A204

CMPDIL Report (1998, April). Operational plan for Jayant

opencast mine, pp 2–20

Gupta RK (1981). Dragline productivity in opencast mines.

Project report, Department of Mining Engineering, BHU,

Varanasi, unpublished

Kishore N (2004). Planning of tandem dragline operations in

opencast mines. PhD thesis, Department of Mining

Engineering, Banaras Hindu University, Varanasi

Lal M, Agrawal MR, Rai RL (1988). Suggested method of

working for Jayant opencast mine operation of dragline in

tandem. In: National symposium on mine planning and

design. BHU, Varanasi, pp 1.01–1.10

Mathur SP (1999) Coal mining in India. M.S. Enterprises,

Bilaspur, pp 96–105

Naganna M, Rai P (2003) Swing time study on large capacity

draglines. CMTM J Indian Inst Coal Manage 8(3):2–5

Rai VK (1997). Balancing diagram—a basic planning tool for

dragline operation. In: Proceedings of the national semi-

nar on emerging technology in surface mines and envi-

ronmental challenges, Mangalore, India, pp 85–89

Rai P (2004) Performance assessment of draglines in opencast

mines. Indian J Eng Mater Sci 11:493–498

504 Geotech Geol Eng (2011) 29:493–504

123