Predictive Analytics for Logistics to Increase Man & Machine Efficiency Richard Martens MD Chicago, 13 th May 2014
Resource Intelligence Self-learning Analysis & Prediction Efficiency increase for man & machinery Discovery: Automatic generation of optimization and simulation model for your specific process Individual: The system propose you an individual solution for the efficiency increasing tightly consider the specific of Your individual approaches or processes Adaptive: Should your processes change in the future, the model adapts automatically to your process If you have questions and their answers are hidden, cut in pieces and incomplete in your data then, Resource Intelligence finds the answers for you! 2
Prof. A.-W. Scheer IS Predict & Scheer Group 800 400 Employees 2010-2014 Visionary, researcher and author of standard works for business information systems Member of the council for innovation and growth of the German Government President of the German Association for Information Technology (BITKOM 2007-2011) Ranked as 2nd most important German IT person (of 100) by Computerwoche magazin (after Hasso Plattner / SAP) Founder of international software companies IDS Scheer & IMC AG Sole Shareholder of Scheer Group GmbH 100 50 Germany Australia Austria Benelux Locations France Great Britain Rumania Switzerland Turnover (million ) 2010-2014 Turkey Ukraine 3
Capacity Planning Reduce Costs for Rental Rail Wagons Optimal allocation of required rail wagon / types Objective: Problem: Solution: Data: Reduce costs for hired rail wagons due to predictive and demand-oriented inventory management Many types of rail wagons difficult to plan best fit wagons in entire context Volatile demand difficult to predict required wagons per category Discovery of patterns and dynamic profiles Wagon demand (history); wagon hierarchy; wagon prices Cost reduction potential due to foresighted-optimal inventory management and allocation of best-fit rail wagon 4
Capacity Planning Reduce Costs for Rental Rail Wagons Optimal Fleet Structure Objective: Problem: Solution: Data: Optimize fleet structure. Fleet hiring to be done in advance without correct and foresighted know how about future demand Discover patterns and dynamic profiles for real future demand Wagon demand (history); wagon hierarchy; wagon prices Cost reduction due to optimal fleet structure planning 5
Plannable Energy Consumption for Fleet of Electrical Vehicles Individual and reliable prediction of charging status to plan range Objective: Problem: Solution: Assurance for driver how far he will get with battery capacity Also considering that battery to be charged with renewable energy, generated by himself No reliable information about charging status available. Thus, no reliable information about possible range. E-car charging status depends on various factors: Driver s behavior, way of driving (sportive, ), usage of electricity consumers (air con, ), distance (stop & go, motorway, ) Dynamic correlation and causality analysis for relevant influencing factors; Highly accurate prediction of relevant factors and charging status Prediction Actual Deviation Charging status Temperature Amperage Torque Highly accurate prediction of relevant individual factors Thus, also highly reliable prediction of battery demand per individual driver 6
Predictive Maintenance Early Discovery of Engine Anomalies Individual demand-oriented maintenance via anomaly analysis Objective: Increase efficiency via early information on (future) wear & tear 10 minutes: Engine run without disturbances Solution: Discover first and hidden signs when engine does not run efficient anymore; inform technical service when thresholds of anomalies is passed 10 minutes: 51 disturbances due to breaks Condition: Individual & cost-reduced analysis per machine without additional sensors Problem: Strongly volatile energy demand, only engine energy data, no access to production data Anomaly Details: No regularity in variable energy demand during disturbance Despite high volatility and no knowledge about production data, engine anomalies are discovered between 86% - 100% - only on energy consumption data! 7
Utilities PV Power Usage Reduced power costs due to optimal usage of own PV power Objective: Problem: How: Data: Run production machinery mostly on PV power, generated by your own PV power very volatile and difficult to plan; energy demand of machinery also volatile; energy demand does not match energy availability Foresighted machinery control via accurate PV power generation prediction Weather (past / forecast) power generation (past) Accurate 24 h PV power generation prediction for 1 individual installation Accuracy Mar Apr May Jun Jul Aug Sep Oct O Month 94 % 97 % 94 % 93 % 99 % 96 % 97 % 92 % O Day 91 % 93 % 92 % 93 % 95 % 95 % 93 % 93 % Resource Intelligence realizes flexible and precise predictions despite high volatility 8
Energy Reduced costs for energy via more precise 24 h gas prediction Objective: Problem: How: Data: Plan demand-oriented gas purchase for tomorrow & thus, reduce purchasing costs Standard load profiles too inflexible for dynamic demand of consumer Dynamic load profiles with flexible pattern recognition Historic gas consumptions, weather (past and forecast); no consumer classification Resource Intelligence April Accuracy Jan Feb Mar Apr May Jun O 24h (%) 96 89 91 88 86 88 Typical Solution Accuracy Jan Feb Mar Apr May Jun O 24h (%) 92 90 81 83 67 74 Resource Intelligence ca. twice as precise than state of the art solution with standard load profiles 9
Why IS Predict? Benchmark on prediction tools in high volatility Common approach Specific forecast External supplier with 12 years experience in area Deviation Average: 18% Maximum: 47 % Discovery Self learning prediction with automatic model generation Deviation Average: 8% Maximum: 26 % 10
SAP CEO visits IS Predict CeBIT 2014 SAP CEO Jim Hagemann Snabe meets IS Predict MD Britta Hilt to inform himself about Resource Intelligence
ff Contact: Richard.Martens@ispredict.com IS Predict GmbH Scheer Tower Uni Campus Nord D5.1 66123 Saarbrücken Germany Phone +49 681 96777-200, Fax +49 681 96777-222 www.ispredict.com 12