OTC-30596-MS

A New Automated Cutting Volumes Model Enhances Drilling Efficiency

Mohammed Murif Al Rubaii, Saudi Aramco D&W; Rahul N. Gajbhiye, KFUPM; Abdullah S. Alyami, Saudi Aramco;
Gunther Glatz and Bashirul Haq, KFUPM; Abdulwahab S. Aljohar, Saudi Aramco; Ala S. AlDogail, KFUPM

Copyright 2020, Offshore Technology Conference

This paper was prepared for presentation at the Offshore Technology Conference originally scheduled to be held in Houston, TX, USA, 4-7 May 2020. Due to COVID-19
the physical event was not held. The official proceedings were published online on 4 May 2020.

This paper was selected for presentation by an OTC program committee following review of information contained in an abstract submitted by the author(s). Contents of
the paper have not been reviewed by the Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect any
position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Offshore Technology Conference is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of OTC copyright.

Abstract
Cuttings concentration in annulus that are generated while drilling due to the rate of penetration have several
problems if it is high and has exceeded the maximum limit. Cuttings accumulation in annulus can lead to hole
problems such as lost circulation coincident, stuck pipe incidents, and slow drilling rate which still a difficult
challenge once it comes to plan, design and drill wells. If a proper hole cleaning efficiency can be achieved,
that ultimately will enable the drilling team to have satisfied well drilling performance. To empower hole-
cleaning performance, it must be engineered. In this paper, a new real time model of cuttings concentration
in annulus (CCA) enhances drilling performance that will ensure optimized improvement and avoid stuck
pipe problems. Knowledge from this paper will help in modeling and monitoring cuttings concentration
in annulus while drilling precisely and therefore facilitate improving ROP without jeopardizing the well
drilling performance. In addition, the paper spoke about the environment impact, drilling waste management
and economics and wellbore instability related to cuttings concentration in annulus.
Huge number of models, techniques, charts, chemicals, tools, methods, and designs, experimental studies
and to enhance the hole cleaning, but these things are just based on theory, lack proper experimental data,
not compatible with drilling scenarios and operations. Cutting Concentration in Annulus (CCA) can provide
a clue or a knowledge for maximum ROP that is save and compatible with rig equipment limitations such
as the knowledge about maximum amount of cuttings generated while drilling that can be transported to
surface and that the shale shakers can handle without causing hole troubles. The size of hole sections (OH),
ROP, flow rate of mud pump (GPM), annular Velocity, Critical velocity (Vc), Drill pipe size (OD), cuttings
rise velocity (Vcr), slip cuttings velocity (Vsc) and transport ratio (TR) properties in certain hole sections
were collected and analyzed first to determine the effect of them on hole cleaning and ROP performance.
The data selected are from the same hole size, formation type and mud type.
The relationships between the collected properties and CCA were t evaluated to determine how strong it
is and demonstrate the significance of properties on estimating CCA. CCA was monitored controlled and
evaluated to be able to have improved hole cleaning performance to enhancement of the drilling operations
and decrease well drilling time. This is the first time to for hole cleaning optimization to lead for ROP
improvement by using real time CCA model. The developed model has been validated using field data
2 OTC-30596-MS

during drilling hole sections. It has shown high drilling rate performance in the hole sections tested and
helped mitigate stuck pipe incidents, increased ROP by more than 52%.

Introduction
Cutting Concentration in annulus (CCA) is defined as an indicator of cuttings volumes that are generated
in annulus of drilled hole section due to used rate of penetration (ROP) while drillings. CCA is also can
guide about the smoothness of used ROP with generated drilling cuttings accumulated in annular area of
hole section, hence it can be used as hole cleaning indicator. It tells about the amount of cuttings can be
removed or transported to surface (shale shaker) that is part of mud solid control equipment to clean the
mud from cuttings that were mixed with the drilling fluids (mud). CCA is a tool that can be used to optimize
well drilling performance or ROP.
Optimization of hole cleaning during drilling operation is very important to enhance the drilling rate,
however optimum hole cleaning in drilled hole sections remains a major challenge. Hole cleaning must be
engineered. The Penetration rate is highly dependent on hole cleaning. Insufficient hole cleaning can cause
stuck pipe, decrement in drilling rate, significant drag and torque, Lost Circulation, wellbore instability,
erratic trends in ECD, more wiper trips, back reaming, bad quality of cement jobs, bit balling and obviously
increment in the cumulative cost of drilling operations and extension of the operations time. If no attention
paid to hole cleaning, such problems can finally be a root cause of losing the well.
Cutting concentration in the annulus is an effective tool that can indicate how much cuttings generated
while drilling, are loaded in annulus. Newitt et al (1955) calculated the cutting concentration in the annulus.
The cutting concentration in annulus or cuttings volume has a limit that is not supposed to be exceeded.
The limit of the CCA is within the range of 5% to 8 %.
If the CCA exceeds the limit, it can strongly lead to severe hole problems. There are several logical
reasons that can explain why exceeding the limit can induce hole problems, see table-1. CCA can help in
optimizing the rate of penetration since the limit is known and recognized. CCA is used to identify that
the optimized ROP that drilling rate can be increased to save or limit without jeopardizing well drilling
performance. CCA has been identified in the literature domain that the save limit applicable the drilling
performance or ROP can be increased up to CCA equals to 5 %, which is applicable to all rig mud solid
control equipment and ensure save and proper optimized performance.

Table 1—Related hole Section problems if CCA is not controlled or optimized.

Problems Impact

Increased PV, YP, Gels), and 6 and 3 RPM readings • Poor cutting transport

• High ECDs,

• Possible break down of formation and lost circulation

Increased fluids loss/thick filter cake • Differential sticking

• High torque and drag

Slow ROP • Chip hold down pressure

Increase in density • Possible break down of formation

• Increase dilution and addition of chemicals to maintain proper

density

Poor cement displacement • Channels that allow pressure communication up the wellbore.

Mitchell (1955) suggested a new method to calculate cutting concentration in annulus. It accounts for
cessation of circulation during connection and circulation which occurs prior to a connection but after
OTC-30596-MS 3

drilling has ceased. This latter circulation is called Pre-Connection Circulation time. Mitchell (1955) stated
in his book (Advanced Oil Drilling Engineering) that American petroleum Institute (API) developed a
new model to calculate the cutting concentration in annulus by using the transport ratio. The same factors
which affect hole cleaning in a vertical hole section also affect hole cleaning in a deviated and horizontal
hole sections. The single major effect on hole cleaning in deviated and horizontal sections is the angle of
inclination. The inclination will form cuttings bed and increase the settling tendency of drilling cuttings.
Cuttings beds will increase the volumetric drilling cuttings concentration in annulus.
There are several factors that can have deep impact on hole cleaning in vertical hole which include the
following:

• Pipe Rotation (RPM)

• Rate of penetration (ROP).

• Flow rate/annular velocity.

• Drilling-fluid density.

• Drilling-fluid rheology.

• Cutting size, shape and density.

Rishi (2000) ranked these factors of hole cleaning during drilling, see figure - 1.

Figure 1—Key Variables Controlling Cuttings Transport-[Rishi 2000].

Drill Pipe Rotation is important to change flow regime from laminar to turbulence. In vertical hole section
the drill string is not eccentric in the wellbore. That will lead drilling cuttings to migrate to the other side
of the wall of wellbore. Laminar flow has maximum velocity in the center. It is recommended to rotate
the drilling string to move drilling cuttings to center to enable the flow rate to lift them easily as shown
in figure - 2.
4 OTC-30596-MS

Figure 2—Pipe Rotation Effect On Drilling Cuttings In Vertical Hole. (Williams 1951).

Drilling rate has a major impact on cuttings transport and hole cleaning. As the drilling rate increases, the
drilling cuttings volume load in the annulus also increases as shown in figure - 3. For efficient transport of
cuttings and hole cleaning ROP should be controlled. Fast drilling, generates more drilling cuttings which
may lead to drilling cuttings accumulation in the annulus and shale shakers especially when drilling across
sticky and clay lithology of formation.

Figure 3—Relationship Between Cuttings Volume in Annulus and Drilling Rate.

A reduction in ROP may have an influential impact on drilling costs, however, the benefit of avoiding
other drilling problems, such as mechanical pipe sticking or excessive torque and drag, can compensate for
the loss in ROP. Drilling rate can be controlled by minimizing cuttings concentration in annulus and critical
velocity, see figure – 4
OTC-30596-MS 5

Figure 4—Relationship Between Critical Velocity and Drilling Rate.

Flow rate is an influential factor that affects hole cleaning efficiency in a vertical hole section. Flow rate
helps the lifting force through momentum transfer and friction because the mud strikes and slips past the
drilling cuttings. Mud weight influences hole cleaning in three manners, provides buoyancy effect to help
lift the cuttings, affects the momentum of the fluid and affects the friction the fluid that can impact drilling
cuttings once they pass through it.
Mud rheology represents the heart of the drilling fluid and provides extraordinary support in the case
of rig equipment pumping limitation. Also, it plays a significant role of leading to perfect hole cleaning.
The size, shape and density of drilling cuttings affect mainly the slip velocity. However, to determine the
previous mentioned features of drilling cuttings with a perfect accuracy is impossible.
If the formation has shale caving or sloughing shale the cutting density, size and shape will be so difficult
to figure out. They are completely different and quite hard to determine them precisely see the figure – 5.

Figure 5—Hole Problem Shale Caving. (Taken from the Rig for 16'' vertical hole section).

There are several factors affecting the hole cleaning in deviated and horizontal sections. The most
important factors are hole angle, cuttings bed, pipe rotation, mud properties and time of circulation. These
important factors are extremely relative to each other.

• Angle of inclination.
6 OTC-30596-MS

• Cuttings beds.

• Pipe Rotation and time of circulation.

• Mud Properties.

• Rate of penetration (ROP) and flow rate/annular velocity.

There are three different zones of inclination in horizontal well, see figure – 6.

Figure 6—Three Zones of Inclination.

Mitchel (2001) stated that the moderate angle (30 – 60) deg. of hole section is the hardest to clean because
of the combination of sliding beds, Boycott settling and asymmetrical flow profile see figure - 7. Cutting
concentration increase dramatically between 30 deg. and 45 deg. and remains relatively constant at higher
angles, see figure - 8.

Figure 7—Hole Cleaning Difficulties. Vs. Hole Angle.

OTC-30596-MS 7

Figure 8—Cuttings Concentration Climbs Rapidly After 30 Deg. (Mitchel 2001).

Mud weight contributes to hole cleaning in higher angles by slowing down the boycott settling effect
and by causing cuttings bed to be more fluidized and less compacted. The cutting concentration in the
annulus decreases as mud weight increases. Baker and Azar (1985) demonstrated the effect of mud weight
on cuttings bed formation. They summarized their findings as follows: Cuttings concentration increase
drastically between 35 deg and 45 deg at low mud weights but not so drastically at higher mud weights, see
figure - 9. Cuttings bed height reduced with small increase in mud weight at any angle. Sliding beds and
avalanching of cuttings beds happened less frequently with heavier mud.

Figure 9—Effect of Mud Weight on Cuttings Bed Height. (Mitchel 2001).

Cuttings beds are more fluidized in heavier mud and thus more easily disturbed. Minimum velocity
needed to initiate cuttings moving is less with heavier mud.
Cuttings beds are formed during the period of low or no pipe rotation that is the case of directional drilling.
Mud properties, flow rate, pipe rotation and hole angle are the major factors that contribute to enhanced
cuttings beds removal. Cuttings bed decreases linearly with flow rate due to the eroding to cuttings beds
by the annular velocity, see figure - 10.
8 OTC-30596-MS

Figure 10—Effect Of Flow Rate On Cuttings Beds. (Mitchel 2001).

Rotation is the key factor in hole cleaning efficiency for high angle holes. It provides active flow area at
the top of hole. Since pipe and cuttings lay along the bottom of hole, the mechanical agitation is required
to get cuttings into the fluid flow. The required rotary speed depends on hole size and rate of penetration
(ROP). Critical pipe rotation should lead to have very good hole cleaning performance see-figure - 11.

Figure 11—Effect Of Pipe Rotation In Hole Cleaning. (Mitchel 2001).

The time to effectively clean the wellbore increases as the angle increases. Flow rate can remove cuttings
even without rotation, but the key here is to have sufficient time of circulation see figures-12 & 13.
OTC-30596-MS 9

Figure 12—Cuttings Beds Erosion Curve For Variable Flow Rates. (Adari 2000).

Figure 13—Cuttings Bed Erosion Curves For Required Circulation Time. (Adari 2000).

Methodology and Results

Cuttings concentration in annulus or drilling cuttings volume is highly dependent on several factors such as
ROP, flow rate of used mud pump of the rig in gallon per minute (GPM), Transport ratio (TR), and the hole
section, critical velocity, slip velocity of the cuttings through the drilling fluid of mud and annular velocity.
Newitt (1955) used all the previous mentioned factors to come up with a model that can estimate or predict
the cuttings concentrations in annulus. After using the model of Newitt and validate it with API model to
estimate the cuttings concentration the results showed in the figure – 14.
10 OTC-30596-MS

Figure 14—Cuttings Volume of Newttis Model VS cuttings concentration of API Model.

The developed model of cutting concentration in annulus used while drilling by utilizing drilling sensors
such as ROP sensor and GPM sensor as real-time acquired data from them, and they can be utilized to
estimate the developed CCA model. The developed model has been validated with API model of estimating
cuttings volume and no need to validate it again with Newitt model. The developed model is defined as a
model that can predict and estimate the cuttings volume in annulus by utilizing drilling sensors of rig.
The developed model has been developed by utilizing based-physics model and use rig sensors. API
cuttings volume model has TR term, which contains the effect of fluid rheology that is represented by
mud properties such as plastic viscosity (PV) and yield point (YP). The rheological properties of drilling
fluid used to estimate the slip cuttings velocity to be calculated and used in TR model, which is (1 minus
slip velocity over annular velocity). To come up with new developed, collecting information is highly
importance, such as mud rheological properties (PV) and (YP), drilling parameters (ROP) and (GPM),
mud type were collected for different filed, hole sections types (vertical, deviated and horizontal), and
formation lithology. The relationships between CCA, and mud rheological properties, drilling parameters
were determined to recognize the most important factors influence CCA.
ROP has major effect on CCA and GPM as well as shown in figure - 15. The methodology to come up
with the developed model was to collect data and utilized rig sensors of ROP and GPM with adding the
effect of PV and YP to consider the rheological influence and recognize the mud type wither, water base
mud (WBM), Oil based mud (OBM) or water with gel sweep. Cutting rise velocity contributed to increase
the cuttings a accumulation if it is increased and that can be described as more cuttings concentration in
annulus would have to increase the cutting rise velocity by adjusting the mud properties or annulus velocity
or controlling ROP to a certain limit that the cuttings concentration can be raised, see figure - 16 for more
clarifications. The transport ratio (TR) has moderated influence on the cuttings volumes or concentration
in annulus. CCA will increase exponentially and that requires more transport ratio to be capable to lift the
cuttings, hence that will allow for some proper optimization to increase the ROP to save and suitable limit
without inducing hole cleaning problems.
OTC-30596-MS 11

Figure 15—Cuttings Volume VS Rate of Penetration (ROP) ft/hr.

Figure 16—Cuttings Volume VS Cuttings rise velocity (ft/min).

CCA is plotted against TR as can be seen in figure – 17. In order to help TR to have efficient ability to
carry out the drilling cutting were generated while drilling circulation time plays an important role to perform
a transport ratio performance optimally. More time of circulation to be spent to transport the cuttings, the
less cuttings accumulations in annulus remain, see figure – 18. The Developed model has been validated
with CCA (Newitt) and CCA API, then the developed CCA model using rig sensors showed excellent
and accepted results. The regression factor is 0.9514 and 0.9949 with different TR values respectively in
accuracy.
12 OTC-30596-MS

Figure 17—Cuttings Concentration in Annuls VS Transport ratio of drilling fluid.

Figure 18—Cuttings Volume VS Time of Circulation (min).

The results were plotted in figure – 19 and 20 respectively. The model has been applied and validated
with CCA API using real-time data and used while drilling to optimize the well drilling performance as
shown in figure – 21. The developed model was used as a real time model by using rig sensors to estimate
the cuttings concentration in annulus and that was plotted against the drilled depths of real time data. The
model was using CCA API model as well against the drilled depths data and they were plotted both of them
with depths and they showed the results as can be seen in figure – 22.
OTC-30596-MS 13

Figure 19—CCA (Newitt) model VS developed CCA model.

Figure 20—CCA (API) model VS developed CCA model.

14 OTC-30596-MS

Figure 21—CCA API model VS developed CCA model Using Real- Time data.

Figure 22—CCA API model and developed CCA model VS Depths (ft) Real-time.
OTC-30596-MS 15

The model has been used to optimize the drilling efficiency in other hole section and gave indications
about the hole cleaning performance and the proper limit of optimization that the drilling team or crew can
increase the ROP to have a certain amount of drilling cuttings concentration in annulus without exceeding
allowable limit that is considered to be safe and proper optimized limit without inducing any hole sections
or drilling troubles, which is 0.05 or 5% cuttings volume.
The model optimized the well drilling performance to 50% ROP improvement as shown in figure – 23.
The orangey color represented the optimized ROP profile using developed CCA model, while the blue
color represented the ROP profile of offset well without using developed CCA model. The application of
developed CCA model was performed on same bit hydraulics features, drill string design, hole section type
(deviated hole section), same drilling rig equipment limitations.

Figure 23—CCA API model and developed CCA model VS Depths (ft) using Real-time monitoring system.

After deployment of the developed CCA model the drilling engineering and operations teams can
recognize and realize that if the current applied ROP during drilling is having the chance or room of
16 OTC-30596-MS

optimization to increase the well drilling performance to a certain or maximum limit that can be considered
as optimum limit of optimization of well drilling performance from the aspect of generating drilling cuttings
concentration in annulus to limit that can be smooth with used ROP values.

Conclusion
After reaching to excellent, optimum and required results, the new developed model has shown a grate
influence on the hole cleaning performance and can lead definitely at the end to the proper optimization of
rate of penetration. The new model was validated with CCA API model and CCA (Newitt) and has shown
0.99 and 0.95 recpectively in accuracy or regression of R square. Developed CCA model is a dependable
indicator for hole cleaing efficiency and save, proper limit to increase ROP to optimum limit with beautiful
and save certainty. The developed model has contributed significantly for hole cleaning evaluation while
drilling and allow to monitor automated or real time CCA profile, that will guide for better estimation or
intervention to optimize either the rheological properties or control drilling rate or ROP to a certain limit
that will allow the mud to transport the cuttings eaier and faster. The developed CCA model will permit
the drilling team to recognize the criticality of hole cleaing optimization for performing enhancement if the
hole sections dictates. The developed model led to optimize the well drilling performance to 55 % in ROP
improvement or enhancement.

LIST OF ABBREVIATIONS
CCA Cutting concentrations in Annulus, %
GPM The flow rate in gallon per mints.
ROP Rate of penetration, (ft/hr)
Vann The annular velocity, ft/min.
Vc The critical velocity, ft/sec.
DP Drill-pipe outer diameter (in)
OH Open Hole diameter (in)
Vcr The cutting rise velocity (ft/min)
Vsc The cutting slip velocity (ft/min)
TR Transport Ratio (%)

LIST OF Equations
1.

2.

3.

4.

5.
6.
OTC-30596-MS 17

7.
(Newtii's Method)
8.
(API Method)
9.
(Automated developed Model)

References
1. Williams, C. E.Jr, And G. H. Bruce. "Carrying Capacity of Drilling Muds." Journal of Petroleum
Technology 3.04 (1951): 111–120. DOI: 10.2118/951111-G.
2. Newitt, D. M, 1955, "Advanced Oil drilling engineering", a text book published by the society of
petroleum engineering.
3. Mitchell, B.J., 1955, "Advanced Oil drilling engineering", a text book published by the society of
petroleum engineering.
4. Maurer, W. C. "The" Perfect-Cleaning" Theory of Rotary Drilling." Journal of Petroleum
Technology 14.11 (1962): 1–270. DOI: 10.2118/408-PA.
5. Becker, T. E., J. J. Azar, and S. S. Okrajni. "Correlations of Mud Rheological Properties with
Cuttings-Transport Performance in Directional Drilling." Spe Drilling Engineering 6.01 (1991):
16–24. DOI: 10.2118/19535-PA.
6. Rasi, Marco. "Hole Cleaning in Large, High-Angle Wellbores." SPE/IADC Drilling Conference.
Society of Petroleum Engineers, 1994. DOI: 10.2118/27464-MS.
7. Luo, Yuejin, P. A. Bern, and B. D. Chambers. "Simple Charts to Determine Hole Cleaning
Requirements in Deviated Wells." SPE/IADC Drilling Conference. Society of Petroleum
Engineers, 1994. DOI: 10.2118/27486-MS.
8. F. E. Beck, Arco Alaska, The Effect of Rheology on Rate of Penetration, Copyright 1995, SPE/
IADC Drilling Conference, Amsterdam, 25 February-2 March 1995. DOI: 10.2118/29368-MS.
9. Leon Robinson and Mark Morgan, 2004 "Effect of Hole Cleaning on Drilling Rate
Performanc"e, Paper Aade-05-Df-Ho-41.
10. Unegbu Celestine Tobenna, Hole Cleaning and Hydraulics, Thesis, 2010.
11. Al Rubaii, M. M. 2018. A New Robust Approach for Hole Cleaning to Improve Rate of
Penetration. Presented at the SPE Kingdom of Saudi Arabia Annual Technical Symposium and
Exhibition, 23-26 April. SPE-192223-MS. https://doi.org/10.2118/192223-MS.
12. Adari, R. B., Miska, S., Kuru, E., and Bern, P. 2000. Selecting Drilling Fluid Properties and
Flow Rates for Effective HoleCleaning in High-Angle and Horizontal Wells. Presented at
the SPE Annual Technical Conference and Exhibition, Dallas, Texas, 1-4 October. https://
doi.org/10.2118/63050-MS.
13. Al Rubaii, M. M. 2017. The Impact of Hole Cleaning on Rate of Penetration. MS thesis, King
Fahd University of Petroleumand Minerals, Dhahran, Saudi Arabia (December 2017).
14. Al Rubaii, M. M., Sehsah, O. R., & Omini, E. (2018, August 16). Approach to Improve
WellDrilling Performance. Society of Petroleum Engineers. doi:10.2118/194223-MS.
15. Al Rubaii, M. M., Al Yami, A., & Omini, E. (2019, March 15). A Robust Correlation
ImprovesWell Drilling Performance. Society of Petroleum Engineers. doi:10.2118/195062-MS.
16. Al Rubaii, M. M., Shehri, H. H., Nwachukwu, C. N., Andrews, S., & Atallah, Y. (2016,
April 25). First Successful Installation of a 4-1/2 Pre-Perforated Liner through a 5-1/2 Solid
Expandable Liner. Society of Petroleum Engineers. doi:10.2118/182814-MS.
18 OTC-30596-MS

17. Al-Rubaii, Mohammed Murif, Al-Abduljabbar, A. M., Hossain, M. E., & Gharbi, S. A. (2018,
January 29). Optimization of Tripping Speed to Minimize Surge & Swab Pressure. Society of
Petroleum Engineers. doi:10.2118/189331-MS.
18. Al Rubaii, M. M. (2018, August 16). A New Robust Approach for Hole Cleaning to Improve Rate
of Penetration. Society of Petroleum Engineers. doi:10.2118/192223-MS.
19. Al-Rubaii, M. M., Gajbhiye, R. N., Al-Yami, A., Alshalan, M., & Al-Awami, M. B. (2020,
January 13). Automated Evaluation of Hole Cleaning Efficiency While Drilling Improves Rate of
Penetration. International Petroleum Technology Conference. doi:10.2523/IPTC-19809-MS
20. Onwukwe, S. I., & Nwakaudu, M. S. (2012). Drilling wastes generation and management
approach. International Journal of Environmental Science and Development, 3(3), 252.
21. Biltayib. M. Dr., 2 Eng. Mufazzal. S. Kabuli., 3 Dania issa, 4 Norah Ayyadand 5 Daliris Ramires
(2016). Management of Drilling Cuttings in Term of Volume and Economics in Oil Field. Int.
Journal of Engineering Research and Application October 2016, pp.45–47
22. Marana, A. N., Chiachia, G., Guilherme, I. R., Papa, J. P., Miura, K., Ferreira, M. V., & Torres,
F. (2009, September). An intelligent system for petroleum well drilling cutting analysis. In 2009
International Conference on Adaptive and Intelligent Systems (pp. 37–42). IEEE.