Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
INTRO TO MANAGEMENT SUPPORT SYSTEMS IS 340 BY CHANDRA S. AMARAVADI IN THIS PRESENTATION.. Introduction to MSS Decisions & types of decisions DSS EIS GDSS 2 INTRO TO MSS 3 INTRODUCTION (FYI) More competition Globalization Complexity More decision making (D.M) 4 MANAGEMENT SUPPORT SYSTEMS MSS: collection of tools/systems to support managerial activity. Characteristics (FYI): Interactive Customizable Model based Support rather than automate 5 MANAGEMENT SUPPORT SYSTEMS ES GDSS TP Reporting DSS EIS AI DSS Evolution Data Mining MSS Note: ES – Expert Systems, AI – Artificial Intelligence EIS – Executive Information Systems; DSS – Decision Support Systems 6 EXAMPLES OF DECISIONS Whether to approve a loan? Whether to promote an employee? How much of an increase to allocate to employees? Where to advertise? Allocation to media? How to finance a capital expansion project? How much to produce? When to produce? What products to produce? What markets? What production techniques to use? 7 TYPES OF DECISIONS When to produce? What products? Types of Decisions Structured problem (routine) Unstructured problem (non-routine) 8 DECISION MAKING STYLES Unstructured Structured D.M. Styles Analytical {focus on methods & models} Intuitive {focus on cues, trial & error} 9 THE IDC MODEL OF DECISION MAKING Intelligence Design Choice Decision ! 10 THE IDC MODEL OF DECISION MAKING Introduced by Herbert Simon, the IDC consists of The following stages: Intelligence -- Identification of problem information Design -- Identification of alternative solutions Choice -- Choosing a solution which optimizes D.M. criteria 11 DECISION SUPPORT SYSTEMS 12 DECISION SUPPORT SYSTEMS A system that supports structured and semi-structured decision making by managers in their own personalized way. 13 CLASSICAL DSS ARCHITECTURE Dialog management User interface Model management Capabilities for creating & linking models Data management Capabilities for managing & accessing data Database Note: model is an abstract representation of a problem 14 DSS ANALYSIS CAPABILITIES “What - if “ Sensitivity Goal-seeking Optimization 15 DSS ANALYSIS CAPABILITIES What if - change one or more variables Sensitivity - change one variable Goal seeking - finding a solution to satisfy constraints Optimization- find best solution under a given set of constraints 16 DSS MODELS (FYI) Financial e.g. portfolio, NPV Statistical e.g. : forecasting Marketing e.g. : product mix, advertising Production e.g. capacity planning, inventory Simulation e.g. production process, bank tellers etc. 17 BANK EXAMPLE Tellers Que1 Que2 Tellers Tellers Que3 Arrival of Customers Waiting Customers Que4 Departure of Customers 18 SIMULATION MODEL Customer Arrives PURPOSE: Identify # of tellers needed, service time Joins Que Is processed Customer leaves 19 CASE OF THE S.S. KUNIANG (FYI) Ship ran aground Owners wanted to sell it Coast guard was the authority Sealed bid Scrap value ($5m) Repair cost ($15m) 20 NEW ENGLAND ELECTRIC SYSTEM Utility company needs coal 4m tons/year Purchased a $70m General Dynamics vessel Capacity 36,250 tons (self loading) Bid for Kuniang? How much? 21 DECISION COMPLICATIONS Type of coal: Egypt or PA? Jones Act and round trip time Exception to Jones Act Self unloader reduces cargo capacity Buy a sister vessel? Tug barge? 22 DECISION OPTIONS (FYI) Options are Kuniang (w crane), Kuniang (no crane), General dynamics vessel, or tug barge 23 DATA FOR THE 4 OPTIONS (FYI) General Dynamics Tug Barge Kuniang Kuniang (Gearless) (Self-loader) Capital cost $70 mil. $32 mil Bid+$15mil Bid+$36mil Capacity 36,250 tons 30,000 tons 45,750 tons 40,000 tons Round trip (coal) 5.15 days 7.15 days 8.18 days 5.39 days Round trip (Egypt) 79 days 134 days 90 days 84 days Operating cost/day $18,670 $12,000 $23,000 $24,300 Fixed cost/day $2,400 $2,400 $2,400 $2,700 Revenue/trip coal $304,500 $222,000 $329,400 $336,000 Revenue/trip Egypt $2,540,000 $2,100,000 $3,570,000 $2,800,000 24 DECISION TREE OF HOW MUCH TO BID Decision Outcome 0.7 Salvage=scrap 0.5 Win ? Salvage=bid Bid $7mil Total Cost NPV Self-Unloader 43 -1.35 Gearless Self-Unloader 22 5.8 43 -1.35 Gearless 28 3.2 Sister Ship 2.1 Tug/Barge -0.6 Lose Note: NPV calculations are based on projections from previous slide 25 NEES ended up bidding $6.7 million for the Kuniang, but lost to a bid of $10 million Coast Guard valued ship as scrap metal Decision tree a useful tool; parameters unknown 26 DSS APPLICATIONS Cash forecasting Fire-fighting Portfolio selection Evaluate lending risk Event scheduling School location Police beat 27 DATA MINING 28 DATA MINING Search for relationships and global patterns that exist in large databases but are hidden in the vast amounts of data. e.g. sequence/association, classification, and clustering 29 Predicting the probability of default for consumer loans Predicting audience response to TV advertisements Predicting the probability that a cancer patient will respond to radiation therapy. Predicting the probability that an offshore well is going to produce oil 30 Associations activities/purchases that occur together e.g. bread and jam. Sequence Activities which occur after each other e.g. car and loan Classification An analysis to group data into classes e.g. pepsi and coke drinkers 31 BI SYSTEMS (ALSO EXECUTIVE INFORMATION SYSTEMS) 32 BI SYSTEMS & DASHBOARDS BI System: Systems that provide information to executives on the business environment. Executive Dashboard: An interface that displays information needed to effectively run an enterprise. Does more information lead to better quality decisions? 33 BI ARCHITECTURE Medline FedStats BI Workstation OLAP/ WAREHOUSE Costs: $50,000 - $100,000 Development time: about 1 month Internal Databases 34 BI CHARACTERISTICS An intuitive easy-to-navigate graphical display A logical structure for easy access Little or no user training is required Data displays that can be customized Regular and frequent automatic updates of dashboard information Information from multiple sources, departments, or markets can be viewed simultaneously EXAMPLES EXAMPLES.. COLLABORATIVE SYSTEMS (GDSS) 38 COLLABORATIVE SYSTEMS An interactive computer based system which facilitates solution of unstructured problems by a set of D.M. working together as a group. Other terms - GDSS, Electronic Meeting Systems. 39 CURRENT BUSINESS TRENDS (FYI) More competition Shift towards flat/virtual organizations More mergers [industry consolidations] Globalization of markets and products More strategic alliances Group D.M. Is it necessary for org. decisions to be made in groups? Why cannot it be handled by individuals? 40 CHARACTERISTICS OF GROUP D.M. Participants of equal rank 5-20 Time limits Requires knowledge from participants 41 A GROUP DECISION SUPPORT SYSTEM Screen Database Org Memory A GDSS System A repository of the D.M. process. 42 GROUP DECISION SUPPORT SYSTEMS 43 GDSS THEORY Process losses - GDSS + Process gains A GDSS minimizes process losses and maximizes process gains 44 ADVANTAGES OF GDSS Time Anonymity Democratic participation Satisfaction Record of decision 45 THE END 46