Mist 420 Chapter 1 summarize .pdf

What are data analytics? Data analytics is a process that involves: (a) identifying the problem (b) gathering relevant data that frequently are not in a usable form (c) cleaning up the data to make them usable (d) loading them into data storage models (e) manipulating data to discover trends and pat...

What are data analytics? Data analytics is a process that involves: (a) identifying the problem (b) gathering relevant data that frequently are not in a usable form (c) cleaning up the data to make them usable (d) loading them into data storage models (e) manipulating data to discover trends and patterns that leads to actionable insights (f) making decisions based on those insights. From data to insights to decisions, data analytics enables us to answer the following questions: What has happened in the past? Why did it happen? What could happen in the future? With what certainty? What actions can we take now to support or prevent certain events from happening in the future? Can some of the actions resulting from our insights be automated? Can the analytics process be automated? Data are the raw gures, numbers, or text that serve as the starting point of analysis. An example of the type of data that businesses typically analyze is sales revenue for each customer. Data become information when they reveal the specific causes or results. For example, we could process sales revenue data to reveal which customers did not make any purchases from us within a given time period. When information is given meaning, we gain knowledge and understanding of the data. Knowledge is created when we learn from information; for example, which customers did not buy from us because of pricing and which customers did not buy from us because of quality? What has happened in Why did it happen? the past? Knowledge knowledge then provides us with the capabilities to gain wisdom. Wisdom, is acquired when knowledge is gathered over time. For example, by examining the reasons why customers do not buy from us, over time we can gain the wisdom to identify and implement policies our company should pursue to acquire and retain customers. Knowledge Knowledge + Knowledge + Knowledge + Wisdom Knowledge+ Knowledge+ Knowledge+ Knowledge What could happen in What actions can we the future? take? Wisdom coupled with a goal yields a decision. For instance, if our goal is to retain customers rather than seeking new customers, then we might launch loyalty programs Wisdom Decision Goal The relationship among analytics in three areas: statistics, computer science, and domain knowledge Statistics is a branch of mathematics that deals with understanding data. It involves the collection, sampling, organization, modeling, analysis, interpretation, and presentation of data. Probability theory plays an important role in statistical inference. Computer science is the study of how computers work and the application of theory to improve computing methods and capabilities. Domain knowledge relates to the expertise gained by individuals in certain areas or elds. For example, medicine is a domain. Data analytics is the intersection of computer science, statistics and domain knowledge. Data science involves the use of computers to acquire knowledge by analyzing large amounts of data using models and domain expertise. Data analytics applies the algorithms and models generated from data science. Therefore, analytics is more concerned with applying models than with creating them. This makes data analytics a subset of data science. The researchers focus on overlap of statistics and domain knowledge. For instance, A medical scientist might use statistical analysis to study the factors that influence whether an individual develops lung cancer. Machine learning is at the intersection of computer science and statistics. Machine learning is the use of computing machines to learn using mathematics model. An example of machine learning is training a computer to learn language structures in order to autocomplete words (and sentences) in mobile applications. The intersection of domain knowledge and computer science leads to the creation of specific software tools that assist in the analysis of specific problems. An example of such a tool is software that helps a radiologist identify tumors in a brain scan. WHY STUDY DATA ANALYTICS? Digital transformation is the transformation of business processes, data, operations, marketing and competition to leverage digital technologies. Data analytics plays a central role in this transformation. Employers of digitally transformed companies seek those that are experienced in data analytics. The skills of data interpretation can be categorized into the following three areas of responsibility: (1) The data scientist is the most specialized analyst, possessing advanced math and computing skills. (2) Data analysts specialize in analysis, and they come from various educational backgrounds. ey typically have an advanced college degree in business, usually with an emphasis in quantitative methods, and data modeling and manipulation. (3) Significantly, most data analysis is being performed by managers and other business users, who may have not received formal education related to data analysis. Examples of how analytics is applied in various fields: Retail: Analytics is used in pricing, timing of pricing strategies, and amount of discounts; product placement; and upselling and cross-selling of products. Manufacturing: A type of data analysis called demand forecasting is the core of manufacturing planning. Marketing: Targeted marketing is made possible with predictive analytics. Marketers can analyze each customer’s behavior and use these data to predict future behaviors. They can then launch targeted individualized marketing campaigns based on this information. Supply Chain: Selecting suppliers and optimizing distribution costs both utilize data analytics techniques. Customer Service/Help Desk: Customized customer service is based on analysis of prior work orders or help tickets, the procedures that succeeded, problem-solving metrics. Forecasting and Budgeting: Most businesses create forecasts and budgets based on historical data and knowledge of the business environment. The use of large quantities of data and analytic tools helps improve forecasting and budgeting accuracy. Audit and Analysis of Internal Controls: Data from internal systems are used to analyze risk and determine how well systems comply with management’s policy of internal controls. Government: The federal government collects census data every 10 years. ese data are made available to researchers, policymakers, state and local government agencies, nonprofit organizations, and the general public. The purpose of the census is to gather demographic data that can lead to better resource allocation and assist in the formulation of effective public policies. Utilities: Data analytics assist utility companies in predicting consumer demands for power and managing the supply of power from producers. Financial Investors: Investors sift through the data of numerous companies to determine which are acceptable investments and which are high risk or unacceptable. Scientists: Scientists and researchers in many fields of study frequently collect vast quantities of data that must be analyzed and interpreted. They gather the data during the course of experimentation, simulations, and samplings, sometimes via sensors. Medicine: Analytics has a variety of medical applications. One application is to identify risk factors that lead to chronic diseases. Data from a specific population are collected over time: diet, exercise, ethnicity, gender, age, occupation, family history, and many other attributes. Factors that influence (both positively and negatively) the outcome are determined using an analytical technique called data mining. Medical practitioners then utilize this knowledge to make decisions regarding prevention and treatment. Sports: Coaches can analyze team and player statistics to develop winning strategies. They can also analyze their players’ physiological data to determine whether their performance can be improved. Fraud Prevention: Analytic techniques enable investigators to flag unusual activities for further investigation. After investigators have identified fraud, they can take corrective actions and implement controls to prevent a reoccurrence. A classic example of this process involves credit card fraud, where an unusual transaction on a cardholder’s account can prompt a denial of the transaction or even a hold on the card. Law Enforcement: Law enforcement uses data analysis to identify patterns of crime in order to allocate resources to the areas and populations that are most impacted by criminal activity which helps reduce crime and the cost of enforcement in those areas. Social media platforms: Social media is an effective way to communicate with a vast number of individuals, whether it be advertising, points-of-view, or casual posts about daily life. Unfortunately, not all posts are reliable or true. So-called “fake news” has had real social, political and economic fallout. Data analytics could play a big role in identifying and perhaps eliminating obvious falsehoods. ANALYTICS METHODOLOGY Enablers are the essential components needed for the methodology to work. They include technology, infrastructure, tools, and techniques. People/Users are generally both the creators and the benefactors of analytics activities. User authorizations, internal controls, and training are required to ensure that analytics is used optimally and securely. The analytics methodology includes 10 key steps: 1. Identify goals: The first step is to define the goals of the analytics process. They can be in the form of quantitative measures, business questions, or qualitative descriptions. Examples of outcome goals are: How can I minimize returns from sales deliveries? Which customer brings us the highest profits? Which factors contribute to product mis-use. 2. Gather data: Data gathering is the next step. In the social sciences, data are often gathered from surveys. In the physical sciences, they are gathered through experiments, simulations, and sensors. In the business world, they are commonly gathered from information systems that record all business activity. 3. Design model: In this step, analysts determine how best to analyze the data they have acquired. One simple analysis involves dividing the data into more defined segments to better understand what they are informing. The process is called slicing and dicing. Slice and Dice Analysis is a data exploration technique that allows data to be dissected, viewed from different angles, and analyzed in depth.  People who require more insightful analysis than simple slicing and dicing employ data mining models. Data scientists have created a number of powerful models. In most cases an existing data model can be used. If the data or the analysis goal is unique, however, then it might be necessary to design a new model. Doing so requires strong mathematical and analytical skills. 4. Apply model: After a model is chosen, it is applied to the dataset. After the model is run against the dataset, the results are presented to the user, often in the form of a visualization such as a chart. 5. Review results: Results are reviewed, and in the case of predictive data models, checked against test data. Any deviations should fall within acceptable model parameters. If they do not, then the model is refined or trained again with new data until the desired accuracy is reached. 6. Present findings: Findings from the analysis process can be presented in various ways. Displaying the results in tabular form (rows and columns), is sufficient if the results are not extensive and are easily understood by the viewer. Visualizations are used to display patterns and trends in large datasets. Interactive dashboards bring together multiple visualizations along with end-user controls. Finally, infographics utilize data and images in a sequential narrative to report the results of the analysis. 7. Derive insights: Insights are drawn using experience and domain knowledge. This is why analytics sits at the intersection of computer technology, domain expertise, and statistics. 8. Make decision: The sound analysis complemented by domain knowledge and analytical skills will lead decision-makers to the optimal decision choice. 9. Deploy strategy: Decisions are then converted to strategies that are deployed as actions. 10. Improve: The outcomes of the actions are measured periodically to assess how well they have met the desired goals. The information obtained from the comparison of goals and actual outcomes is used to improve the process in the future. It also serves as an input at the beginning of the next cycle. After the improve step has been completed, we go back to step 1 of the cycle and start the process again. Summary: Data analytics entails gathering data from various sources and making them usable for analysis. The cleaned-up data then need to be loaded into a data storage structure where we can manipulate them to provide insights that assist us in the decision-making process. The demand for individuals with the skills to perform quality data analysis is high and continues to grow as the volumes of data continue to proliferate and organizations increasingly recognize that data analysis is crucial for them to flourish in today’s environment. This chapter highlighted the analytics process, which involves a 10-step cycle that begins with goal setting, but never actually ends because the improvements derived from the results of the analysis process feed back into establishing new system or process goals. Global Bike Company (GB) Company History: John Davis and Peter Weiss are co-CEOs Operations in the United States (US) and Germany (DE) The Business: Professional and prosumer cyclist market GB sells high quality bicycles and cycling accessories Sells to retailer partners, not to the consumer directly GB Business Structure  GB Products Master data are business entities such as Customers, vendors, products, employees, fixed assets…etc. Transactional data are data about an event such as Sales, purchases, pay employees, collect payments…etc. GB uses an integrated system (ERP) for managing operations Some GBI Employees and Their Business Questions Nina Kane – U.S. Sales Manager How do I anticipate and compete with new competitors for market share in the prosumer bike market? What data do I need to evaluate sales, discover trends, and identify opportunities? How can I motivate my sales team? Is there value in organizing invitation-only promotional events for our loyal customers? Are there any delays in our supply chain that have led to lost sales? Donna Vasant – Business Analyst Are there new markets for GBI products? Are there new products that GBI could bring to the market? How about bike-integrated video cameras, body-monitoring devices, and a bike trolley to carry pets, babies, and groceries, especially in urban areas? Is an electric bike a viable and profitable new product? Are internet sales the future for high-end bikes (instead of or in addition to retail stores)? Will internet sales compete with our loyal customers (distributors)? Jessi Mard – Controller Are profit margins in line with expectations? Are operational efficiencies affecting profitability? How can we encourage customers to pay more quickly? How well can we forecast cash flows and other key accounting indicators? Are our internal accounting controls effective? Peter Pollard – Production Manager How can we become more efficient in the manufacturing processes? Is the quality of raw materials affecting production? Can the manufacturing plant meet sales demand? Is the manufacturing facility set up to reduce unnecessary material movements and encourage efficiency in the production process? Tony Liu – Shipping Personnel What percentage of orders are shipping within 24 hours of order receipt? Bruce Hewlett – Shipping Manager How many orders have not shipped within 24 hours of receipt and why? How many products were damaged during the packing and shipping process? Did any orders ship to the incorrect customer or address? How many orders were damaged by the shipping company? How many orders did not arrive at the customer’s site on time due to shipping problems and how can those problems be avoided in the future?