This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: MthSc 810 – Mathematical Programming instructor: Pietro Belotti Martin Hall, O 321 phone: (864) 656 6765 office hrs: Tue/Thu, 3:30pm – 5pm (or by appt.) email: [email protected] web page: http://myweb.clemson.edu/˜pbelott Excerpt from the syllabus Homework: 25% (approx. one every week) Midterm I: 25% (end of September) Midterm II: 25% (end of October) Final exam: 25% Learning and exercising material: ◮ Textbook : D. Bertsimas, J.N. Tsitsiklis, Introduction to Linear Optimization , Athena Scientific. ◮ Extra material: handouts and the online material by Bob Fourer: www.4er.org/CourseNotes ◮ Modeling languages : for modeling optimization problem. Many have a limited version available to students. AMPL: preferred. No Graphical User Interface (GUI), but I know it better (read: I can help) – www.ampl.com Mosel: very nice GUI – google “xpress mosel” GAMS: Has version with even nicer GUI (Aimms) – www.gams.com , www.aimms.com Lecture plan ◮ Relaxations, lower/upper bounds, convexity, complexity ◮ Linear Optimization problems ◮ The simplex method ◮ Duality theory ◮ Sensitivity analysis ◮ Large scale problems: decomposition methods ◮ Network flow models Optimization models ◮ are used to find the best configuration of processes, systems, products, etc. ◮ rely on a theory developed mostly in the past 50 years ◮ when applied in an industrial, financial, military context, they can yield a better use of budget/resources Success stories Source: http://www.informs.com (see also http://www.ScienceOfBetter.org ) yr company result 86 Eletrobras (hydroelectric energy) 43M$ saved 90 Taco Bell (human resources) 7.6M$ saved 92 Harris semicond. prod. planning 50% → 95% orders “on time” 95 GM – Car Rental +50M$ 96 HP printers — redesigned prod. 2x production 99 IBM — supply chain 750M$ saved 00 Syngenta — corn production 5M$ saved An example You are charged with designing a can , a cylinder made of tin (a can is obtained by cutting and soldering tin foil). ◮ The can must contain V = 20 cu.in. (11 fl.oz., 33 cl) ◮ Tin (foil) is expensive, use as little as possible ⇒ Design a cylinder with volume V using as little tin (i.e., total area) as possible. Example r h If we knew radius r and height h , ◮ the volume would be π r 2 h ◮ qty of tin would be 2 π r 2 + 2 π rh π r 2 h must be V = 20in 3 ⇒ h = V π r 2 Rewrite the quantity of tin as Q ( r ) = 2 π r 2 + 2 π r V π r 2 , or Q ( r ) = 2 π r 2 + 2 V r ⇒ Find the minimum of Q ( r ) ! Minimize the quantity of tin 1 2 3 4 5 6 5 0 1 0 0 1 5 0 2 0 0 2 5 0 Q ( r )[ in 2 ] r [in] minimum r = 1.471 in h = V π ( 1 . 471 ) 2 = 2 . 942 in Your first Optimization model Variables r : radius of the can’s base h : height of the can Objective 2 π rh + 2 π r 2 ( minimize ) Constraints π r 2 h = V h > r > Optimization Models, in general, have: Variables (e.g. height and radius, number of trucks, . . . ): the unknown (and desired) part of the problem (one thing...
View
Full
Document
This note was uploaded on 03/14/2012 for the course MTHSC 810 taught by Professor Staff during the Fall '08 term at Clemson.
 Fall '08
 Staff
 Math

Click to edit the document details