Yield Management as A Process Governed by Data Mining in the auto Industry

Information Systems

Yield Management as A Process Governed by Data Mining in the auto IndustryAnalytics, Big Data and the CloudEdmonton, April 23, 2012 Author: Ayman Ammoura M.Sc.OutlineIntroducing main conceptsApplying our science and technology to a Canadian small businessMining on The Revenue Side - RatesMining on The Expense Side InsuranceSharing success stories

Yield ManagementYield management is the process of understanding, anticipating and influencing consumer behavior in order to maximize yield or profits (Wikipedia)Understanding Observation and analysisAnticipate Forecasting Influencing Management actions

Data MiningData Mining is a step in the knowledge discovery process. (Osmar Z.)Data mining is a process of extracting previously unknown, valid, and actionable information from large databases then using the information to make crucial business decisions (Cabena, et al, 1998)Data Warehouse Data repository built to facilitate OLAP (OnLine Analytic Processing) not OLTP (Transaction).Warehouse Multidimensional, Subject-Oriented, data model Data CubeTo support OLAP, a data warehouse is often implemented as a hierarchical N-Dimensional data cube.Data CubeRental DaysLocationTimeVehicle ClassUsually you need SIC, Source, Sold Extras .. N-DimesionsFact TableDimension TableTimeClassLocationEach slice it an n x m 2D TableData Cube Sliced

ProfitabilityThere are 2 items that define the financial well being of an organization.Revenue (our example Rental Days)Expense (our example Insurance)In our case, we need to create a data repository with Fact tables Rental Days and Insured Units

KDD Process: Cleans & TransformThis fires @ 4:00 AM Everyday

Master Control

Daily @ 0600Revenue: Units on Rent

Canada Winter Games Revenue: Fine Detail

Revenue: Rental RatesHow and when to adjust.Utilization Based rate adjustmentNot Competitive Big missed opportunities (explained next)To answer the When question we needed to get more insight into the dataUnderstanding the Cycle

City Sold-outRevenue: Utilization based Tiers Create a system that would issue new booking rates based on utilization.0%- 50% +0%51% - 65% + 10%66% - 75% + 15 % etc This will be transparent to the agent and has been widely used for over a decade.

Rate Control AlgorithmBuild Availability CubeEvery 10 MinutesBranch RatesPublish IntranetWalk-in RatesSystem WideUsing this model, we were able to increase revenue by 30% in the first cycle (May-September)Revenue: Guaranteed CyclesDuring busy season, booking are received 90 days in advanceShoulder Season as low as 6 days average

90 daysSold OutRevenue: Busy Cycle considerationsUsing the utilization tiered rate adjustment process alone 50% of the business can be improved by at lease 20% Because 50% booking is required to achieve the next tierOn Average, most bookings during busy cycle were entered 3 months in advance

Rate Control Algorithm IIBuild Availability CubeEvery 10 MinutesBranch RatesPublish IntranetWalk-in RatesSystem WideInsert Cyclical Adjustments

Known DatesRevenue: Result SummaryUp $1.3MillionUtilization based TiersUp $2.2MillionUtilization + Cyclical and Localized AdjustmentsPhase I and Phase II were constructed one cycle apartComplete project spanned 14 monthsNext ExpenseSo far we talked about an example of how we applied simple Data Mining tools to achieve great results on the revenue side, helping a small business. Next we will examine how we have effectively used analytics to impact profitability by reducing a major expense.Expense: Insurance AnalysisNext to depreciation, this is usually the second biggest expense in the auto industry.Existing Scenario is that the business had to pay the insurance premium per unit ($m) on all used units in a calendar month. Existing solution was: Identify units that were rented (n), and pay monthly ($mxn)How to reduce this cost?Insurance: Activity Analysis

Visualization of the number of active days of every insured unit for a typical monthInsurance and UtilizationExamining the number of insured units against the number of units on rent

Insured VehiclesRented VehiclesCountExpense: Insurance As there are more units in the fleet than was required, the company insured way more than was required Information that was implicit dataTime to renegotiate the insurance model! Preferably without sharing your results with the broker

Expense: Result SummaryInsurance cost decreased by $120,000 per yearInstead of paying on all units, we negotiated a policy that allows us to pay higher prorated premiums but on a daily basis. Without the ability to transform the data into information, this effort was unnecessary and probably have not happened!Recall our definition (Data mining is a process of extracting previously unknown, valid, and actionable information)

Thank You Love to answer any questions .