PART A: Using Descriptive Analytics to address a business need:

The vehicle groups identified by horsepower and fuel efficiency (in city and highways mpg) have shown four major clusters, different characteristics as well as preferences in the driveline.

Cluster ID Number of Rows Number of Rows (Ratio) Within-Cluster Sum of Squares Fuel_Info.City_mpg Fuel_Info.Motorway_mpg Engine_Stats.Horsepower

1 1,223 0.24602696535 1,797.4608552892 1.357251974 1.1884375596 -1.1209059857

2 812 0.16334741501 710.9467622253 -0.73574079397 -0.52032880423 1.6381000253

3 1,094 0.22007644337 639.6493273038 -1.0667308755 -1.0501003109 0.25166133883

4 1,842 0.37054918527 842.8908782065 0.056735116384 0.0639383175419 -0.12735434569

Excluded Rows 0 0 0

Cluster 1 is about high horsepower cars with moderate to high fuel economy on the motorways suggesting that their preference tends toward too powerful vehicles, despite their relatively high efficiency and likely suitability to long distance journeys. Moreover, this bunch of values (24.6 % of the data set) indicates a consumer group, which is looking for a combination of performance and efficiency.

Area 2 is represented by 20.3% of the cars here but with an impeccably lower horsepower and

favourable city mpg on the contrary. With these cars, we can cover this said priority of fuel

efficiency in urban stop-and-traffic where power in traffic is unworthy regarding.

For Cluster 3, moderate power and minimal fuel utilization in both scenarios are observed to be

the mixing attributes. This implies that are hanging older models or not optimized ones, as well as

those that are heavy on power and don’t reach the high horsepower within Cluster1.

The largest group of vehicles with an engine power limited to 37.1% and good fuel economy in

the city especially comes in cluster 4. This segment is the mass-market cluster, which is aimed at

customers, placing environment-friendly technologies above power, and presenting a smart

combination of both urban and lengthy road trips.

The pictures do not dispute the above findings as they provide the distribution curves that indicate

obvious differences in the horsepower and mpg categories of these groups. The biplot PCA which

projects influencing these differences even more has the clusters distributed along the principal

components that have a clear distinction between high horsepower and high fuel efficiency.

Therefore, individual attributes of vehicles will be considered and assessed, as well as the liking

and desires of car owners.

PART B: Using Predictive Analytics to assist with a business problem.

Based on the trend analysis of oil and gas stocks from 2003 to 2022, the sequence underscores performance behaviour of stock over a time span, indicating illustrative insights for strategic investment purposes. Chevron (CVX) and ExxonMobil (XOM) feature tendency to a close to the top tracking the regression line implying that these companies have established leadership and are recognized by the market as safe investments.

Symbol 2.. 2.. 2.. 2.. 2.. 2.. 2.. 2.. 2.. 2..

Year of Date

Close

COP

CVX

MPC

MRO

PSX

SLB

VLO

XOM

Sum of Close (size) broken down by Date Year vs. Symbol.

On the flip side, Marathon Oil (MRO) and Schlumberger's (SLB) charts reveal higher volatilities and generally average close prices hence showing riskiness or different dynamics in the market that affects these stocks.

The data lines prove that all stocks show a sharp decline in stock prices around 2020, which may probably relate to the worldwide pandemic of coronavirus (COVID-19) that had indeed a deep impact on global fuel demand and prices that had led to the decline of petroleum. While the trading patterns of those two companies have showed an almost immediate recovery after oil prices corrected, other companies such as BP have not demonstrated the same speed in their recovery, indicating a weaker resilience or a fundamental adjustment to changing market conditions.

The trend of sum of Close for Date Year. Color shows details about Symbol.

COP

CVX

MPC

MRO

PSX

SLB

VLO

XOM

Based on the statistics, investment is expected to harvest the best recoveries during the slumps, like the observations that investors should explore selling at the peaks as in 2014 and 2019 which could lead to the maximization of profits. Primarily, historical trends could solely guide investment decisions without having into count other indicators like broader economic variables and future market forecasts thus calculations becoming extremely wrong. In this regard, a well-calibrated perspective which embraces current market analysis, and future predictions should be the subsistence of investment strategies in the oil and gas business--an industry with its volatility.

PART C: Example of Prescriptive Analytics

Advanced businesses identified prescriptive analytics as a revolutionary approach to management, introducing it as a personalized, smooth, and high-performance solution to capitalize on the best opportunities. A well-illustrated example is given through a logistics company that, by using a technology-driven approach to their complex capacity problem, known as "kernelized empirical risk minimization" solved, accomplished, their capacity problem in a simple manner. They took up machine learning algorithms to predict the capacities based on historical demand data, thus showing unprecedented evolutions of performance compared to traditional approaches (Notz & Pibernik, 2022).

Real-time decision support system (DSS) implementation --in the form of a manufacturing company of specialty steel bars that is in North America -- is the last example of benefit. This DSS is intended to support allocation of available inventory and improve ATP decisions. This system employed mixed-integer programming models as a tool to facilitate quick and efficient decision-making processes which greatly improved the company's data-based operational capacity (Mahdavi Pajouh et al., 2013).

Reflection:

In short, prescriptive analytics as a new business tool entirely permits forecasting of the future set of events with specific modelling approaches and simultaneously suggests adapted business practices that are in line with the existing organizational goals and objectives that eventually result in high performance and gain competitive edge.

PART D: Storytelling

Of all the libraries across Europe, the resources they were all availed tells a story of cultural diversity and the diverse priorities of public investments. Italy is clearly the frontrunner boasting 12,435 libraries (22,5 million readers) which demonstrates a wide availability of libraries in society. This significant total of libraries can, in turn, lead to high participation of readers demonstrating accessibility also supports their use.

Sum of Total Libraries for each Country.

In 2015, France with 5,242 libraries counted around 12.8 million users gives the proof of public top services in culture and education, likely reinforced by government support and a deeply rooted literary tradition.

Denmark and Estonia might be less evident examples but considering this a little differently may provide a different perspective. Danish institutions cater for about 2.5 million users, which means they have impressive per capita utilization that could most probably be the result of effective outreach as well as well-planned services that focus on community needs. Estonia, that is split across 977 libraries, for the total of 853,177 users, presents the picture of setting and accomplishing the library services in the smaller nations.

Total Libraries and Total Users for each Country. Color shows details about Total Libraries and Total Users.

Final thoughts

This representation is illustrating that cultural, economic, and governmental factors assume more

important role in laying down the foundation for library resources and their application across

the world and as a result the educational landscape and access to the knowledge will be

completely different in different countries.

REFERENCES:

Mahdavi Pajouh, F., Xing, D., Zhou, Y., Hariharan, S., Balasundaram, B., Liu, T., & Sharda, R. (2013). A specialty steel bar company uses analytics to determine available-to-promise dates. Interfaces, 43(6), 503-517.

Notz, P. M., & Pibernik, R. (2022). Prescriptive analytics for flexible capacity management.

Management Science, 68(3), 1756-1775.