Mining Goes Digital – Mueller et al. (Eds)
© 2019 Taylor & Francis Group, London, ISBN 978-0-367-33604-2
367
Short-term production scheduling of multiple mines
using genetic algorithms
P. Pathak & B. Samanta
Indian Institute of Technology Kharagpur, Kharagpur, West Bengal, India
ABSTRACT: Short-term production scheduling of a mine determines the extraction
sequence of blocks on a shorter scale of months, weeks or days. It models the mine operations in
great detail and it typically aims to meet the quality and quantity targets of production in each
period. Over the years, a number of models have been developed to optimize the short-term
production scheduling problem of the mines. However, most of these models were developed
for single mines. This paper presents a model to optimize short-term production schedules of
mineral value chains comprising of multiple mines. In addition, a genetic algorithm based heuristic approach has been presented to tackle industrial scale instances of short-term production
scheduling problem which exact solution techniques cannot solve within a reasonable amount
of time. Computational experiments suggest that the presented approach is a promising way to
handle complex industrial scale instances of short-term production scheduling problem.
1 INTRODUCTION
Short-term production scheduling plays a pivotal role in optimizing the operations of a mine. It
is carried out on a shorter scale with a granularity that typically spans from a week to a month
(at maximum) and a horizon which spans from few months to a year (Blom et al. 2018). It seeks
to find out, for each short period, the combination of mining blocks which should be extracted,
the time at which they should be extracted and the destinations at which they should be sent
such that the quality and quantity requirements are met consistently subject to the technical
and operational constraints of production. Often, the short-term production scheduling is carried out under the scope of long-term production scheduling and thus the former also plays a
key role in linking the operational level activities of a mine to its strategic intent.
Researchers, over the years, have developed a number of models to optimize the short-term
production scheduling problem (Sundar & Acharya 1995, Smith 1998, Askari_Nasab et al.
2011, Eivazy & Askari-Nasab 2012, Bakhtavar & Oraee 2012, Mousavi et al. 2014, Kumar &
Chatterjee 2016, Matamoros & Dimitrakopoulos 2016, Mousavi et al. 2016, Morales & Reyes
2016). However, most of these models were developed for mineral value chains comprising of
a single mine. Also, majority of these models were solved through exact solution techniques
which can only handle small scale instances of the problem and are unable to solve industrial
scale problems within a reasonable amount of time (Mousavi et al. 2014).
In this study, a generalized mathematical formulation to optimize the short-term production scheduling of mineral value chains comprising of multiple mines has been presented.
The formulation nearly models all the relevant technical, practical and operational aspects of
short-term production scheduling problem. In addition, a genetic algorithm based heuristic
has also been designed to tackle industrial scale instances of the problem.
2 MATHEMATICAL MODEL
The short-term production scheduling problem of mineral value chains comprising of multiple mines has been formulated as a mixed integer linear programming problem (MILP).
© 2019 Taylor & Francis Group, London, ISBN 978-0-367-33604-2
367
Short-term production scheduling of multiple mines
using genetic algorithms
P. Pathak & B. Samanta
Indian Institute of Technology Kharagpur, Kharagpur, West Bengal, India
ABSTRACT: Short-term production scheduling of a mine determines the extraction
sequence of blocks on a shorter scale of months, weeks or days. It models the mine operations in
great detail and it typically aims to meet the quality and quantity targets of production in each
period. Over the years, a number of models have been developed to optimize the short-term
production scheduling problem of the mines. However, most of these models were developed
for single mines. This paper presents a model to optimize short-term production schedules of
mineral value chains comprising of multiple mines. In addition, a genetic algorithm based heuristic approach has been presented to tackle industrial scale instances of short-term production
scheduling problem which exact solution techniques cannot solve within a reasonable amount
of time. Computational experiments suggest that the presented approach is a promising way to
handle complex industrial scale instances of short-term production scheduling problem.
1 INTRODUCTION
Short-term production scheduling plays a pivotal role in optimizing the operations of a mine. It
is carried out on a shorter scale with a granularity that typically spans from a week to a month
(at maximum) and a horizon which spans from few months to a year (Blom et al. 2018). It seeks
to find out, for each short period, the combination of mining blocks which should be extracted,
the time at which they should be extracted and the destinations at which they should be sent
such that the quality and quantity requirements are met consistently subject to the technical
and operational constraints of production. Often, the short-term production scheduling is carried out under the scope of long-term production scheduling and thus the former also plays a
key role in linking the operational level activities of a mine to its strategic intent.
Researchers, over the years, have developed a number of models to optimize the short-term
production scheduling problem (Sundar & Acharya 1995, Smith 1998, Askari_Nasab et al.
2011, Eivazy & Askari-Nasab 2012, Bakhtavar & Oraee 2012, Mousavi et al. 2014, Kumar &
Chatterjee 2016, Matamoros & Dimitrakopoulos 2016, Mousavi et al. 2016, Morales & Reyes
2016). However, most of these models were developed for mineral value chains comprising of
a single mine. Also, majority of these models were solved through exact solution techniques
which can only handle small scale instances of the problem and are unable to solve industrial
scale problems within a reasonable amount of time (Mousavi et al. 2014).
In this study, a generalized mathematical formulation to optimize the short-term production scheduling of mineral value chains comprising of multiple mines has been presented.
The formulation nearly models all the relevant technical, practical and operational aspects of
short-term production scheduling problem. In addition, a genetic algorithm based heuristic
has also been designed to tackle industrial scale instances of the problem.
2 MATHEMATICAL MODEL
The short-term production scheduling problem of mineral value chains comprising of multiple mines has been formulated as a mixed integer linear programming problem (MILP).