||In the past decades, attempts were made to build a systematic approach to mining method selection (MMS) Ooriad et al, (2018). This is because MMS is a complex and irreversible process. Since it can affect the economic potential of a project, the approach must be as thorough, precise, and accurate as possible. The key challenges of the previously established techniques such as the Nicholas and Laubscher method are that, there was a lack of engineering judgement in the process of selecting a mining method. In other instances, not all the parameters required in the mining method selection process were considered; i.e. economics would be the basis of the final decision of a mining method without taking into consideration other factors such as geology (Bogdanovic et al, 2012). While other techniques just considered a few parameters and a limited number of mining methods as alternatives (Namin, 2008). Some techniques were customised procedures for a specific orebody (Namin et al, 2009). Each orebody is unique; therefore, the approach of just adopting the same mining method for similar commodities was not always an effective or realistic approach Therefore, the existing procedures were found to be inadequate and not applicable for consideration in all MMS processes.
To solve the challenges stated above, an up-to-date approach to MMS is the use of multi-criteria decision-making (MCDM) tools to aid in the process. The MCDM are effective in facilitating a decision-making process; however, the use of MCDM has not gained enough popularity across countries and in the mining industry especially in MMS (Mardani et al, 2015). Their successful implementation in other industries such as in manufacturing companies, water management, quality control, transportation, and product design (Lee et al, 2007)present an opportunity for further exploration in MMS. In this research, these MCDMs were further explored as starting point to solving the challenge faced in MMS.
With the aim of developing a systematic and an unbiased approach that caters for subjective and objective analysis in MMS, this study investigated 10 MCDMs- TOPSIS, TODIM, VIKOR, GRA, PROMETHEE, OCRA, ARAS, COPRAS, SAW, and CP with potential to solve the MMS challenge. The study focused on deriving a model where the MCDMs can be integrated and be successfully used for MMS. Included in the research are factors and mining methods that are necessary MMS. The aim was to use the factors and mining methods as inputs to the developed MMSM.
In the result section, case studies were used to analyse the MCDMs following a descriptive and a statistical analysis (sensitivity analysis, spearman correlation, and Kendall’s coefficient.). PROMETHEE, TOPSIS, and TODIM stood out as methods for use in the selection of mining method in the coal mining industry. From the research findings, it was generally concluded that OCRA, ARAS, CP, SAW, and COPRAS are simplified approaches of the afore-mentioned methods. VIKOR’s rankings were outlying and the conclusion was that it was not a suitable method for MMS. GRA’s conclusion based on the literature view was that there remain many unanswered questions about its mathematical foundations.
The MMSM was developed using the results obtained. In the MMSM, first, the user defines the problem. The approach is of case-based reasoning (CBR); where the user can retrieve, re-use, revise and then retain the information (in the database) for future use. The user can always search within the database for a similar problem to select a MCDM, factors and methods; and this may be one of the future areas of improvement on the developed MMSM because there are a number of factors, MCDMs, and mining methods that the user may need
to go through before getting to the relevant MCDM. One of the recommendations made by the author was that the user must understand the theoretical background of the MCDM before using it in the MMSM. In future studies, algorithms for selection of a suitable MCDM in the MMSM can be developed so that once the problem has been defined and structured; the user may not struggle with knowing which method to use amongst the suggested. Also, an application-based approach may be investigated further.