Development of well complexity calculator and its integration into standard well engineering management system/well delivery system

Oil and gas well drilling is the most important and complex task for oil and gas exploration. It is not necessary that design and execution complexity remain the same for two different wells even in the same field. It is possible to have a very complex well to drill after a very straightforward simple well being drilled earlier in the same field. Making correlation or comparison of any of the two or more than two oil and gas drilling wells is an ongoing debate in the petroleum industry. Generally, companies compare the oil and gas drilling wells on a single or two parameters, for example: time versus depth, directional trajectories, well cost and/or other single factors in disengagement of one another. In order to compare two different types of oil and gas drilling wells, having distinctive design, drilling and fluid program and challenges, a scientific rating system is required, which can relate various wells with one another. In this research paper, a calculator named Well Complexity Calculator has been developed to measure the complexity of the oil and gas well drilling by using different parameters. All these parameters are commonly affecting the drilling program and its execution. Secondly, a methodology is designed for integration of Well Complexity Calculator into standard Well Engineering Management System/Well Delivery System for better execution of drilling program. Fifty-one (51) oil and gas drilling well complexity parameters have been utilized to develop Well Complexity Calculator, where they are categorized into three main complexities types named Design Well Complexity, Geological Well Complexity and Project Well Complexity. Design and Geological Well Complexities combine to form Drilling Well Complexity, and then Drilling Well Complexity and Project Well Complexity combine to form Well Complexity. Median, Mode and Monte Carlo simulation techniques were chosen to develop the calculator where Median showed best suited results and was accordingly chosen for the final calculator. Sixty-six (66) actual oil and gas wells’ camouflaged drilling data were used to analyze and fine tune the developed Well Complexity Calculator. Output complexities of these wells were falling in different complexity levels. Moreover, it was seen that the number of low, high and medium complexity wells was different for Design, Geological, Project, Drilling and Well Complexities which is in line with the real-world scenario. The findings and the output Well Complexity Calculator can be very useful at any stage from initial planning to close-out of a well. Without the application of a system like Well Complexity Calculator, wells are categorized as low, medium or high complexity based on either two to three major parameters or based on qualitative assessment of team involved in the project. Here, step-by-step procedure is developed and explained by which any company involved in Drilling and Well Operations can develop their own Well Complexity Calculator and then accordingly integrate it into their Well Engineering Management System/Well Delivery System.


Introduction
Vulnerability is enormous in drilling operations and activities, particularly in Exploratory Wells or the wells with limited subsurface data. Drilling and Well Operations comprise numerous unpredictable operations and steps, which include both controllable and non-controllable elements/factors. Every one of these variables add to the complexity of the well and likewise sufficient possibilities should be worked in time and cost assessments of the drilling activities . In any case, the most ideal approach to correlate various wells is by ranking them while considering maximum possible parameters. Well Complexity Calculator developed here considers multiple factors to produce a single digit which is true representative of Well Complexity. Several studies have proposed methodologies and concepts that help quantify the costs and complexity of the well due to different parameters such as constraints posed by well drilling and planning factors, broadly classified by Well Complexity Index. It was proposed that many researches have been carried out to determine the well complexity and its costs; however, reservations exist when evaluating and predicting the factors which influence the performance of the well operations (Kaiser 2007). A number of conventional techniques have also been proposed in this aspect, where offset records are used to analyze the actual performance of the well in an oil field; however, these standards and records are subjective in nature and tends to be viable for short range and technical performance standards (Pessier and Fear 2013).
According to , the complexity of well operations, including drilling, completion, testing and stimulation of the well, is a risk-prone process that requires accurate amalgamation of scientific and technical concepts of petroleum engineering. However, these mathematical and computational concepts for precise calculations are integrated with the consideration of far more subjective and qualitative well operations which includes the physical hazards such as environmental catastrophe. This poses a continuous hazard throughout the well operations.
To safeguard the economics and safety of the well operations, a number of studies show the incorporation of various factors which need to be considered during the contingency planning of the complex wells which are difficult to be drilled and require a competitive budget strategy (Ezenwanne and Giadom 2018; Curry et al. 2013;Oag and Williams 2013;Dupriest 2013). Mason and Judzis (2013) proposed that the method to calculate the future performance of future higher step-out wells are very challenging. They considered that departure-to-TVD ratio, often attributed as ERD, explains the difficulty of drilling offshore wells. They demonstrated that the offset well data reveals the "Risks and Limits" of the drilling such difficult wells. The Saudi Arabian Oil Company also mentioned a similar issue by which they classified their ERW as per the measured depth (MD) and derived a new method for designing deep wells that reached a total depth of 17,600 ft (Muñoz et al. 2016).
According to , the complexity of well operations, including drilling, completion, testing and stimulation of the well, is a risk-prone process that requires accurate amalgamation of scientific and technical concepts of petroleum engineering. They developed Drilling Complexity Index (DCI) and Planning Complexity Index (PCI) and accordingly declared their combination as Well Complexity Index (WCI). They declared it as is a very useful tool which categorizes the complexity of the well. It is used as contingency indicator for the level of complexity to be faced in the well operations. DCI can be used at any interval of well delivery process. DCI is often used in planning of new wells. It is suggested by researchers that DCI is used in predicting the non-productive time (NPT) of the drilling operations, time and cost of the drilling plans, and portfolio management and found Drilling Complexity Index has found to be of major importance in predicting the hazards and risks in the overall well operations . Drilling Complexity Index (DCI), together with Planning Complexity Index (PCI), is classified as Well Complexity Index (WCI). DCI involves the challenges of subsurface operations, such as drilling rigs and equipment. Planning complexity index covers surface challenges, considering geopolitics of the well location, the 1 3 climate and well logistics. These two factors are very important in project evaluation of well planning and operations .
Drilling a well is a very expensive venture, and several factors influence its drilling cost; the common characteristics are considered to be the rent of the drilling rig, transport and weather conditions, etc. Irrespective of the high-tech equipment, the weather downtime always incorporates considerable variation in the time required to drill a well (Jenkins and Crockford 2013). Loberg et al. (2013) considers the economic feasibility an important aspect in well planning, alongside technical considerations. Probabilistic well cost estimate was suggested that would strengthen and systemize the corresponding workflows of the drilling of the well.
It is suggested that Well Complexity Index is a very useful technique that allows the comprehensive planning and operating the drilling projects, addressing a broad range of aspects, from the well planning phase to the drilling phase. WCI significantly helps to evaluate the resources needed for the project and anticipated non-productive time ). In any case, the most ideal approach to correlate various wells is by ranking them while considering maximum possible parameters. In this research paper, Well Complexity Calculator (WCC) has been developed where 51 oil and gas drilling well complexity parameters have been utilized to develop Well Complexity Calculator. The parameters selected here are more in quantity compared to work already done in this regard .
The parameters which are used to develop in previously published work are very few. Using a few parameters to define oil and gas well drilling complexities sometimes results in erroneous outcomes. In previous work, the detail of the procedure for application of the Well Complexity Index was also not available. In contrast, in this research paper, parameters are categorized into three main complexities types named Design Well Complexity, Geological Well Complexity and Project Well Complexity. Design and Geological Well Complexities combine to form Drilling Well Complexity, and then Drilling Well Complexity and Project Well Complexity combine to form Well Complexity. This categorization is different from the work already done in this regard ) and allows to see the impact of design-related parameters, geological parameters and project-related parameters separately as well as in combination (Fig. 1).
Earlier work also doesn't explain the process of preparation of the calculator itself. Here, step-by-step procedure is presented following which any Company involved in Drilling & Well Operations can develop their own Well Complexity Calculator and accordingly integrate it into its Well Engineering Management System/Well Delivery System.
The later part of this paper consists of different sections. Firstly, materials and methods are presented in Sect. 2, which represent the main parameters selection and calculator development methodology. Section 3 is results and discussion and the last section, which is Sect. 4, is Summary and Conclusions.

Development of well complexity calculator: materials and method
The main major material required in this research study is to have the data of different oil and gas drilling wells, related with complexities of drilling at different levels.

Formulation of well complexity parameters and sub parameters
Based on the literature review and consultation with different industry experts, multiple parameters were formulated under each category; considering the application of Well Complexity Calculator on the single lateral Wells being drilled only in the onshore environment. Table 1 presents the parameters for each category used for the development of Well Complexity Calculator. These parameters are more in quantity compared to work already done in this regard ).

Formulation of sub-parameters
All the parameters mentioned in Table 1 were further divided into the sub-parameters. Sub-parameters against each parameter are listed in detail in Appendix B.

Categorization of well complexity
Work already performed on Well Complexity Indices involves either single major category or fewer categories . Here Well Complexity was categorized into following three main categories: 1. Design well complexity 2. Geological well complexity 3. Project well complexity However, based on the combined effects of three main categories mentioned above, following two categories are produced: 4. Drilling well complexity (combined effect of Sr # 1 and 2) 5. Well complexity (combined effect of Sr # 1, 2 and 3)

Development of survey forms and conducting the survey
After formulating Well Complexity Parameters and Sub-Parameters, different options were considered for getting the weightage factor against each parameter and rating against each sub-parameter. Workshop was one of the options, but it was not possible to gather all the relevant industry experts and professionals in one workshop for this purpose. Therefore, option of online survey was selected being easier and convenient to get the input from all the industry experts and professionals. Google Survey Forms were utilized for this purpose. Extract from the Google Survey being used here is presented in Appendix A.
Aim of the survey was to rate different parameters against each other. Linear numeric rating scale from 1 to 10 was provided for recording the responses. Numeric 10 means that this parameter has the strongest impact on complexity, whereas 01 means that this particular parameter has the weakest impact on the complexity. Then sub-parameters were to be rated within each parameter separately. Most complicated/difficult sub-parameter was awarded 10, whereas most easy sub-parameter was awarded 00. Other sub-parameters within the parameter were awarded from 10

Data quality check
After receiving raw data from survey, a rigorous exercise has been done for the data quality check. Aim of the data quality check was to filter the survey results which were ambiguous or were not in line with the question being asked. On certain instances, it was observed that survey participants could not understand the complexity parameters and hence entered the same rating against each of the sub-parameter. Such types of answers were removed from the survey in order to make the quality of the results better and more reliable.

Analysis of survey results
After quality check of the raw survey data, different types of analysis were performed on the data. Analysis was done mainly on following three methodologies. It is pertinent to mention that mean/average of the data produced very misleading results and hence was not adopted for the analysis.
1. Median. It is a value or quantity lying at the midpoint of a frequency distribution of observed values or quantities, such that there is an equal probability of falling above or below it. 2. Mode. It is the number which appears most often in set of numbers. 3. Monte Carlo Simulation. There are many ways in which Monte Carlo Simulation can be defined. However, it is a broad class of computational algorithms that rely on repeated random sampling, standard deviation, mean and probability to obtain numerical results. Accuracy of Monte Carlo Simulation depends on the number of iterations which are selected, however increase of iterations can increase the time required for each calculation.
Here in this analysis, 5000 iterations were selected with 55% probability, which has showed reasonable consistency in the results.

Normalization and compilation of analysis results
After performing the quality check and three analysis mentioned above, obtained results were normalized. Aim of the normalization was to make the highest rated sub-parameter within each parameter equal to 10 and the lowest rated equal to 0 and accordingly rest of the sub-parameters spreading as per the ratio of actual survey results obtained. This was done to make the results in line with the original aim of the Liner hanger application 20 CwD/RwC application 21 Under-reaming/bi-center bit application 22 Cementing operations 23 New technology application survey, which was not obtained as such due to the inputs from the survey participants. Following equation was used to normalize the survey results.
Thereafter, resultant ratings of each sub-parameters were multiplied with the weightage factor of each well complexity parameter to produce the end results of each well complexity sub-parameter. After applying weightage factor of parameters on the rating of each sub-parameter, resultant values made all the sub-parameters quantitatively comparable with each other although many of them are purely qualitative in nature. This normalization and compilation of survey results was performed for all the three analysis methodologies (Median, Mode and Monte Carlo Simulation). Figures 2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51 and 52 present all the three methodologies results of Well Complexity Sub-Parameters for each parameter. Out of the three methodologies, Median showed the best suited results for all the parameters without any abnormality in the trend lines/graphs, whereas Mode and Monte Carlo Simulation results show some abnormality in the trend lines on some instances which was primarily due to the type of survey input and not because of the methods itself. Anomaly in Monte Carlo result is obtained in Fig. 2, whereas anomaly in Mode result is obtained in Fig. 31. Median has not showed anomaly for any result. Accordingly, Median results were selected for onward analysis and utilization for

Creation of well complexity calculator in excel spreadsheet
Based on the results listed in Appendix-B, Well Complexity Calculator was created in Excel Spreadsheet with the simple user interface. User is required to input the well information and select sub-parameter against each parameter for the well under investigation. Screenshot of the Well Complexity Calculator is presented in Appendix C. All the parameters are created with the drop-down option to select one of the available sub-parameters, whereas Parameters # 22, 24, 28 and 32 are with multiple selections of sub-parameters. However, each of the available sub-parameter cannot be selected in this case as well, for example in Parameter # 24, out of "Single Loss Zone" and "Multiple Loss Zone" one sub-parameter can be selected and likewise out of "Single Gain/Influx Zone" and "Multiple Gain/Influx Zones" one sub-parameter can be selected to avoid erroneous results. Hence, calculator is macro-enabled,        and VB codes are included in it in order to apply these logics which restrict user to select one of these two sub-parameters while allowing the selection of other sub-parameters.
Based on the sub-parameters selected by user, scores are picked which are accumulated and accordingly Well Complexities were calculated on the Scale of 10 through Eqs. 2-6.         After calculating Well Complexities of all the wells, it was observed that Eqs. 2-6 need adjustment with some factor since denominator in these equations was leading to comparatively lower values for all the five Well Complexities. Hence, sensitivity was run and it was concluded that factor of 0.75 in denominator gave the resultant Well Complexities similar to prima-facie complexities of these Wells.
This factor doesn't change the complexity metric scale and is introduced to align the calculated Well Complexities with the perceived complexities of the well ). This factor would vary depending on the type of wells in hand. Accordingly, based on the actual wells data presented here, Well Complexity Calculator equations were adjusted and are presented from Eqs. 7-11.

Divisions of well complexities
Based on the Well Complexities calculated for actual Wells, following three division are concluded. However,

Integration of well complexity calculator in standard well engineering management system/ well delivery system
Review of standard well engineering management system/ well delivery system.
Well Engineering Management System/Well Delivery System of any company is a structured and a step-by-step approach for planning, execution and close-out of any drilling project

GATE -IV
System provides the means to capture lessons learned and to retain knowledge for future reference (De Wardt 2010) (Fig. 58). Without proper Well Engineering Management System/Well Delivery System, well planning, execution and close-out would always go through different approach each time the well is planned, executed and closed out. Different companies use Well Engineering Management System/ Well Delivery System adjusted as per their business methodologies. However, basic concept and structure remains the same which aim toward following the same pattern each time the work is done and to include more steps as the approach becomes maturey.
Basic stages/phases in any Well Engineering Management System/Well Delivery System are as under:

Integration of well complexity calculator in concept & select phase/stage
Concept & Select phase starts with the receiving of well design requirements and criteria from G & G team, based on which well designing is done, whereas normally project specific organogram is prepared based on overall workload distribution, instead of criticality or complexity of the project itself.
In Concept & Select phase, Well Complexity Calculator can be used as the basis of resources/manpower allocation for the project according to its complexity level. Based on the Well Complexity value, project team of relevant experience and expertise can be formulated for a well in hand. Furthermore, during preparation and evaluation of different well design options, Well Complexity value for each well design can be calculated and be one of the key factors in making the selection of well design. Therefore, during the peer review meeting, Well Complexity value for each well design option can be presented along with pros and cons of different well design options in order to have a quantitative perspective of each well design along with conventional approach of qualitative selection.

Integration of well complexity calculator in define & design phase/stage
In Define & Design phase/stage, detailed well designing is performed, followed by more refined Time & Cost estimation which is then used in preparation and approval of Well AFE.
Well Complexity Calculator can be used for more refined time calculation. Normally, time estimation of any well depends on the available information from off-set wells for similar activities in which time for additional planned activities is added on same basis/assumptions. Using Well Complexity Calculator, a more structured approach can be adopted for calculation of time contingency and expected non-productive time by comparing Well Complexity value of a well in hand to the historical data of wells having similar complexity values. Thereafter, expected drilling time estimated using Well Complexity can be directly used to calculate the Well AFE. Moreover, comparison of AFE can be made with the historical data of Well Complexity vs Dry Hole Drilling Cost for further refinement.
In Define & Design phase/stage, the next important utilization of Well Complexity Calculator is to set Key Performance Indicator (KPI) Targets of the well in hand. If there is no Well Complexity Calculator in use, usually this step is

Summary and conclusion
Without the application of a system like Well Complexity Calculator, wells during planning phase are categorized as low, medium or high complexity based on either two to three major parameters or based on qualitative assessment of team involved in the project.
Furthermore, wells during the close-out phase are categorized as low, medium or high complexity based on the actual downtime/problems encountered during execution instead of actual Well complexity. Through the application of Well Complexity Calculator, Well Complexity can be computed considering multiple parameters. The main conclusive outcomes of this study are: The research outcomes from this study also show some limitations, which are: • The limitation of this study is kind of input that it received from the experts carrying out the initial assessment of the parameters and sub-parameters. • This limitation can be eradicated by careful quality check of the input data which has been done here in this research paper. • After rigorous exercise of data quality check, subsequent steps were carried out in order to prepare the Well Complexity Calculator. • Another limitation is the kind of well which was available for the verification of the calculator are single lateral wells being drilled in onshore environment. The calculator developed here is applicable for single lateral wells being drilled in the onshore environment. • However, steps explained in this research papers will remain same for preparation of Well Complexity Calculator with the inclusion of number of laterals and drilling environment as two more parameters; and accordingly this limitation can be eliminated as well.