96 ISEIS Journal of Environmental Informatics Journal of Environmental Informatics 33(2) 96-104 (2019) www.iseis.org/jei Mixed Integer Linear Programming for Oil Sands Production Planning and Tailings Management M. M. Badiozamani 1 , E. Ben-Awuah 2 * , H. Askari-Nasab 1 1 Mining Optimization Laboratory, School of Mining and Petroleum Engineering, Department of Civil and Environmental Engineering, University of Alberta, Edmonton, Alberta, T6G 2W2, Canada 2 Mining Optimization Laboratory, Bharti School of Engineering, Laurentian University, Ramsey Lake Road Sudbury, ON, P3E 2C6, Canada Received 11 November 2015; revised 21 October 2106; accepted 8 January 2017; published online 30 January 2019 ABSTRACT. In oil sands open-pit mining, further processing of the extracted oil sands generates massive volumes of tailings. To save space, the tailings are deposited in in-pit tailings containments constructed by internal dykes using mine waste material. In this paper, an integrated mine planning framework is proposed and implemented using mixed-integer linear programming to optimize the production schedule with respect to dyke construction and in-pit tailings deposition. A case study is carried out to verify the performance of the proposed optimization model. The results demonstrate how the produced tailings are deposited in the excavated mining pits as the mining operation proceeds and the in-pit dykes are constructed using mine waste material. The framework facilitates sustainable oil sands mining through a reduced environmental footprint. Keywords: dyke construction, mine planning, mixed integer linear programming, oil sands mining, production planning, tailings management optimization 1. Introduction An oil sands deposit is a mixture of bitumen and water in sands and clay. Oil sands mining comprises of the removal of overburden material and the mining of the oil-bearing Mc- Murray formation. It is one of the fastest growing industries in North America. Though in recent times oil prices are relatively low, there have been considerable investments in the past that can keep this industry vibrant for some decades. It is also more relevant now that further research aimed at improving the pro- fitability and sustainability of these operations in the long-term is pursued through robust mine planning. An efficient mine plan determines the best schedule for extraction and the destination of the extracted material, in a way that maximizes the net present value of the mining project. In oil sands operations, the material mined is sent to the proce- ssing plant for extraction of bitumen through hot water extra- ction process, which produces tailings. About 80% of the mat- erial sent to the processing plant ends up in the tailings dam. These tailings facilities require large amounts of waste mate- rials for their construction. Oil sands mining operations generate considerable volu- mes of solid waste mostly as overburden and interburden (OI) to access the mineralized zone. The current practice is to dump the waste material for later use mainly in dyke construction and reclamation. The dykes may be constructed either in-pit or ex- pit depending on the waste management strategy in place at the time. The main source of the required material for dyke cons- truction is OI material coming from the mining operations, and the tailings coarse sand (TCS) coming from the processing pla- nt (Fauquier et al., 2009; Ben-Awuah, 2013). Ben-Awuah et al. (2012) provide a detailed description of an integrated oil sands mining operation including material flows and solid waste ma- nagement. Hence, waste disposal, reclamation planning and dyke construction planning can be well integrated with the mi- ne planning framework. In the literature, few works have add- ressed such integration, but none of them has covered the men- tioned domains completely (Ben-Awuah and Askari-Nasab, 2011; Ben-Awuah et al., 2012; Badiozamani and Askari-Nasab, 2012a; Badiozamani and Askari-Nasab, 2012b; Ben-Awuah, 2013; Badiozamani and Askari-Nasab, 2013a; Ben-Awuah et al., 2015). Mathematical programming models such as goal program- ming and mixed integer programming have frequently been used for solving industrial and municipal waste management problems (Ahluwalia and Nema, 2007; Xu et al., 2014). Since the 1960s, operations research techniques in the form of linear programming, integer programming, mixed-integer linear pro- gramming (Johnson, 1969) and dynamic programming (Tan and Ramani, 1992) have been used to find the optimized pattern of extraction and to determine a destination for the extracted material in open-pit mining and block caving (Newman et al., * Corresponding author. Tel.: +1 7057651151; fax: +1 7056754862. E-mail address: ebenawuah@laurentian.ca (E. Ben-Awuah). ISSN: 1726-2135 print/1684-8799 online © 2019 ISEIS All rights reserved. doi: doi:10.3808/jei.201900405