Managed Analytics
and Market Insights

USE CASES

Internval vs. External

General Examples (Cana Advisors)

Improve organizational operations management, planning, and use of data to make decisions.

Craft analytics/operations research-based decision support planning solutions in logistics, maintenance, and supply chain operations.

Enhance organizational data quality, processes, and planning through analytics assessments and action plan execution.

Apply statistical, simulation, and optimization-based tools and analysis to support improvements to decision making.

Establish cost savings and significant investment returns with a focus on improving performance, readiness, and capabilities.

Provide operations research and studies analysis for concept development and employment support course of action analysis and decision making.

DESCRIPTIVE STATISTICS

"What has happened?"

Charts & Graphs

  • Histograms

  • Scatter Plots

Numerical Presentations

  • Mean

  • Median

  • Mode

  • Variance

  • Standard Deviations of Distributions of data

  • Cross tabulations

PREDICTIVE STATISTICS

"What could happen?"

Forecasting (Time-series)

  • Moving averages

  • Auto regression

Simulation

  • Discrete event: a company might simulate the hiring of new employees every year to determine the impact on the company's output. Each year, a different number of employees are hired and each hiring serves as a discrete event in its own right.

  • Monte Carlo: As a simple example, assume that the owner of a rolling hot dog stand earns an average of $75 in profit per day. He sells hot dogs for $3 each, and expects daily sales to vary between 100 and 200 hot dogs. All hot dogs that do not sell spoil and are a total loss, and he puts 200 hot dogs in his stand each morning just in case he has a big day. Using Monte Carlo analysis techniques, the owner could study the effects of various business decisions based on the performance of a key indicator, such as daily profit. For example, he could test the profit impact of the decision to start each day with only 175 hot dogs in his cart. He could then generate a series of random numbers to represent daily hot dog sales over a given period to determine his average daily profits. He may find, for instance, that profits rise when he starts out with only 175 hot dogs, despite the fact that he will run out of hot dogs on some days.

  • Agent-based modeling

Regression (Linear, logistic, step-wise)

Regression is commonly used by businesses to evaluate the effectiveness of their marketing campaigns. For example, if a company funds a campaign for advertisements on both television and radio, it can use regression to judge how their marketing costs are affecting sales. Regression also allows this company to capture the isolated impact of marketing on TV and radio each as well as their combined impact.

Statistical Inferences

  • Confidence Intervals: can predict how many items a business is expected to sell in the next fiscal quarter. Say one of the products is computers; an analysis can be done to give a confidence estimate and range of the number of computers that they will sell for the next 3 months (analysis example: 95% confident that the count of computers sold will be between 78 and 111). The confidence intervals can change based on a desired percentage of confidence, and the business can then prepare things such as their computer stock based on the resulting interval.

  • Hypothesis Testing: another way an organization can evaluate marketing effectiveness is via a hypothesis test. Imagine a company runs a short trial of an advertising campaign for a few weeks. Looking at the sales revenue afterwards, it seems as if sales did increase during the campaign; however, how can the company be sure that this was a result of the marketing rather than random variability of sales? A hypothesis test can be conducted with a chosen degree of confidence to tell if this revenue increase was significant enough to indicate a correlation with the advertising campaign.

  • Analysis of Variance: A car manufacturer needs to determine which steel company to supply the steel for their car parts. The car company receives a large group of samples from each of the 5 different steel companies, and they measure and record the strength of each sample piece of steel. While sample averages can be calculated to compare the steel suppliers, analysis of variance is more useful because it also calculates the variability of these samples. If a supplier’s average steel strength is marginally lower than a competitor’s, it may still be wise to go with them if their steel strength is less variable and more consistent.

  • Design of Experiments

 

Classification

A product seller can use classification to predict whether potential customers will buy their product or not. Using data from their past marketing campaigns and purchases, a model can be developed using demographics (such as age, gender, location) of who did and did not buy their product. The seller can then collect demographic data on potential customers and use the model to predict which ones have a high probability of buying; the company can then target these specific customers while not wasting money or effort marketing towards consumers who are unlikely to buy their product.
 

Clustering

A grocery store can use clustering to divide their customer base into different segments for using different marketing strategies. Purchase data from the store’s checkout system can then divide them into groups depending on what items they most commonly buy; examples of such groups would be “fresh food customers” who often get fruits and vegetables or “junk food customers” who usually purchase chips and other snacks. Once these clusters are created, the store can then run targeted campaigns at these groups that differ to each one’s tendencies.

Artificial Intelligence

Game Theory

  • Prisoners Dilemma: suppose that Company B enters a new market where only one business, Company A, is selling products in. Company A has the choice to either amicably welcome Company B’s competition, which would allow both businesses to keep their pricing around market equilibrium, or aggressively slash their prices to keep all of the now-competitive market. Game Theory is the idea of Company B realizing this choice that Company A has and planning their business strategy around which choice they predict Company A to make.

PRESCRIPTIVE STATISTICS

"What do we do about it?"

Optimization (see solver.com examples)

  • Linear programming: Restaurants use linear programming for menu planning. It optimizes meal production and thereby increases restaurant profits. It uses varying material cost, material supply, and profit from each menu item. This way management can determine the cost of preparing different menu items to decide how many of each menu item to prepare for optimal profit.

  • Integer programming

  • Nonlinear programming

  • Mixed integer programming

  • Network optimization: a shoe manufacturer has 3 production facilities where they create their product and 4 stores where they sell them. They must transport their shoes from the production facilities to the stores with a limited number of truckloads over varying distances between each facility and store. Network optimization shows them how many truckloads should move between each facility and designated stores to minimize fuel and transportation costs while meeting inventory demands at each store.

  • Dynamic programming

  • Metaheuristics

    • Rich portfolio optimization, index tracking, enhanced indexation, credit risk, stock investments, financial project scheduling, option pricing, feature selection, bankruptcy and financial distress prediction, and credit risk assessment.

    • Market basket analysis: checkout recommendations from Amazon

  • Simulation-optimization

  • Stochastic optimization

Examples by Modeling Technique (CAP Examination Guide)

CORPORATE FINANCE

Working Capital Management

Capital Budgeting

Inventory Management

Cash Management

Capacity Planning

DISTRIBUTION

Distribution Model

Multi-Level, Multi-Commodity Distribution Model

Partial Loading

Facility Location

Production/Distribution Model

MERGER & ACQUISITION

Make this your own. Click here to add the content you want.

SALES & MARKETING

(SMART FOX)

Make this your own. Click here to add the content you want.

INVESTMENTS

Portfolio Optimization (Markowitz Model)

Stock Portfolio Management

Portfolio Optimization (Sharpe Model)

Bond Portfolio Management

Bond Portfolio Exact Matching

PURCHASING

Contract Awards

Inventory Stocking/Re-ordering

Media Planning

Purchasing/Distribution Model

SUPPLY CHAIN

Make this your own. Click here to add the content you want.

PRODUCTION

Product Mix

Machine Allocation

Blending

Process Selection

Cutting Stock

HUMAN RESOURCES

Crew Scheduling

Office Assignment

Employee Schedulign

WORKFORCE

Make this your own. Click here to add the content you want.