SlideShare una empresa de Scribd logo
1 de 24
Descargar para leer sin conexión
A study on a simple heuristic for Simple Assembly Line
                 Balancing problem

                     Monard VONG

                      Optimization Lab.
                   Seoul National University


                   December 15th 2009




  M. Vong (SNU)             SALBP              December 15th 2009   1 / 24
Outline




1   Introduction to Assembly Line Problems


2   Solving SALBP


3   Worst case performance ratio and computational experiment




      M. Vong (SNU)               SALBP               December 15th 2009   2 / 24
Concept of Assembly Line (AL)




                                               Figure: Tasks and precedence
                                               constraints
   Figure: Assembly Line Concept

                                                   In assembly line balancing
                                                   problem, the process (tasks
      The amount of time a
                                                   and constraints) is known
      workpiece can be processed by
      a station, before the conveyor               The goal is to find a feasible
      belt moves the workpiece to                  assignment of tasks to
      the next station is called the               stations.
      cycle time.

    M. Vong (SNU)                      SALBP                     December 15th 2009   3 / 24
Simple Assembly Line Balancing Problem
      Now we consider the simplest case of an AL problem: Simple Assembly Line
      Balancing Problem
      The objective is to maximize the efficiency of the line.


Characteristics
  1   paced serial line;
  2   deterministic (and integral) task time tj ;
  3   assignment restrictions: precedence constraints ;


      From now on, we focus on SALBP-1:
           cycle time c is fixed
           minimize the number of stations required
      This problem is NP-hard.


       M. Vong (SNU)                     SALBP                 December 15th 2009   4 / 24
SALBP-1 as an optimization problem

Input
    Set of task V = {1, . . . , n} with task time tj ∈ N, ∀j ∈ V
    Precedence digraph G = (V , A) (partial ordering of tasks)
    Cycle time c

Goal
    Find number of station m to maximize the line efficiency. max tsum
                                                                m·c
    with total task time tsum = n tj
                                 j=1


Output
    S1 , . . . , Sm , m subset of tasks such as ∪k∈1,...,m Sk = V and Sk is
    feasible for all station.


        M. Vong (SNU)                SALBP                  December 15th 2009   5 / 24
Notations


         n          Number of tasks;
         V          Set of task V = 1, . . . , n
         m          Number of stations; index k = 1, . . . , m
         m∗         Optimal number of station
         m, m       Lower, upper bound on m∗
         tj         Task time of task j = 1, . . . , n

         Pj         Set of direct predecessors of task j
         Pj∗        Set of all predecessors of task j
         Fj∗        Set of all followers of task j
         Sk         Set of task assigned to station k: Station load
         t(Sk )     t(Sk ) = j∈Sk tj , k = 1, . . . , m



    M. Vong (SNU)                     SALBP                 December 15th 2009   6 / 24
State of the Art




    Lot of research done on SALBP since 50 years.
    Many heuristics and optimal procedures have been designed.
    Most research is focused on solving the problem with additional
    constraints.
    But there is few research on worst case ratio of heuristic algorithms.
         Queyranne(1985): no polynomial time algorithm achieves an absolute
         worst case performance ratio less than 3
                                                2




     M. Vong (SNU)                 SALBP                  December 15th 2009   7 / 24
Outline




1   Introduction to Assembly Line Problems


2   Solving SALBP


3   Worst case performance ratio and computational experiment




      M. Vong (SNU)               SALBP               December 15th 2009   8 / 24
Construction schemes for heuristic algorithm



    Most existing algorithm enumerates solutions by constructing them
    successively assigning tasks or subset of tasks to stations.
Definition
 Availability A task j is available if all predecessor h ∈ Pj∗ have been
              assigned.
Assignability An available task j is assignable to a station k if the current
              idle time of k is sufficient.
Maximal Station Load A station load Sk is maximal if no available task is
            assignable to k.




      M. Vong (SNU)                 SALBP                  December 15th 2009   9 / 24
Task-Oriented Greedy Heuristic for SALBP-1 Problem



Adaptation of Next-Fit algorithm
 1: topologically sort the tasks
 2: k = 1
 3: for i = 1 → n do
 4:   if t(Sk ) + ti ≤ c then
 5:      Sk ← i
 6:   else
 7:      k ←k +1
 8:   end if
 9: end for




     M. Vong (SNU)                 SALBP   December 15th 2009   10 / 24
For instance c = 10




                           Figure: Precedence Graph




                      Figure: Solution Obtained by Next-Fit

     M. Vong (SNU)                    SALBP                   December 15th 2009   11 / 24
Adaptation of First Fit algorithm
 1: topologically sort the tasks
 2: m ← 1
 3: for i = 1 → n do
 4:   if ∃k ∈ {1, . . . , m} such that
      tSk + tj ≤ c and the precedence constraints are respected then
 5:      Sk ← Sk + {i}
 6:   else
 7:      m ←m+1
 8:   end if
 9: end for




     M. Vong (SNU)                SALBP               December 15th 2009   12 / 24
For instance c = 10




                            Figure: Precedence Graph




               Figure: Solution Obtained by adaptation of First-Fit
     M. Vong (SNU)                    SALBP                  December 15th 2009   13 / 24
c = 10




                     Figure: An optimal Solution with 6 stations


     M. Vong (SNU)                      SALBP                 December 15th 2009   14 / 24
Outline




1   Introduction to Assembly Line Problems


2   Solving SALBP


3   Worst case performance ratio and computational experiment




      M. Vong (SNU)               SALBP               December 15th 2009   15 / 24
Worst case performance ratio



   Notations
        I is an instance of our optimization problem.
        z ∗ (I ) is the optimal value solution.
        z H (I ) is the value of the solution obtained by heuristic H
   The worst case ratio of H is defined as the smallest constant ρH such
   that z H (I ) ≤ ρH z ∗ (I ) for each instance of I ∈ I
   The asymptotic worst case ratio of H is defined as the smallest
   constant ρH for which there exist another constant δ such that
   z H (I ) ≤ ρH z ∗ (I ) + δ for each instance of I ∈ I.




    M. Vong (SNU)                     SALBP                   December 15th 2009   16 / 24
Upper Bound on the worst case ratio


Proposition
An upper bound of the worst case ratio of any heuristics which build maximal
station load is 2.

Proof.
For any heuristic H which build maximal station load, for any station k built by
this heuristic: t(Sk ) + t(Sk+1 ) > c
                                          mH                 n
 mH              mH ·c                          t(Sk )       j=1 tj
 k=1t(Sk ) >      2 .    However, m∗ ≥    k=1
                                            c            =   c        , then we obtain
m < 2 · m∗ .
 H



     Corollary
           The worst case ratio of Next Fit is 2, and it is tight.



       M. Vong (SNU)                     SALBP                             December 15th 2009   17 / 24
A 2 worst case instance for SALBP

   We will provide an instance for which, even the adaptation of First Fit
   algorithm to SALBP cannot beat the 2 worst case factor.
                          1C
   Let consider 0 < δ ≤   6 k:
        2k tasks of task time 3δ,
        2k tasks of task time 1 C − 2δ,
                               2
        2 ∗ (2k − 1) tasks of task time δ




                       Figure: 2 Worst case instance


    M. Vong (SNU)                   SALBP                  December 15th 2009   18 / 24
Figure: Optimal Solution




                Figure: Solution given by the algorithm


                      A(I )         2k
                  lim         = lim      =2
                 k→∞ OPT (I )  k→∞ k + 1

M. Vong (SNU)                    SALBP                    December 15th 2009   19 / 24
Ranked Positional Weight

     Using a topological sorting is not enough, the quality of the sorting may be
     improved.
     Ranking Positional Weight (RPW) of task i is the sum of the task time ti
     and the task time of all followers of i. RPWi = ti + j∈F ∗ tj
                                                               i

     Tasks are sorted by non-increasing RPW.

Example

                                               RPW1 = 38       RPW6 = 18
                                               RPW2 = 23       RPW7 = 17
                                               RPW3 = 32       RPW8 = 13
                                               RPW4 = 27       RPW9 = 11
                                               RPW5 = 22      RPW10 = 2

Sorting by non increasing RPW, the following order is considered.
1, 3, 4, 2, 5, 6, 7, 8, 9, 10
      M. Vong (SNU)                   SALBP                  December 15th 2009   20 / 24
Using RPW as topological sorting, the optimal solution is obtained for the
     previous worst case instance with the same first fit algorithm.

Ranked Positional Weight algorithm
 1: Sort the tasks by non-increasing ranked positional weight
 2: Use First Fit rule

     What about the worst case of the improved algorithm then?




      M. Vong (SNU)                   SALBP                     December 15th 2009   21 / 24
If we used Next fit algorithm, instead of First Fit, it is still possible to find
instances for which the (asymptotic) worst case ratio is 2.
Consider n integer and δ > 0 such that 2nδ < 1.
The instance is made of:
     n tasks of task time 1 C + δ
                          2
     n tasks of task time 1 C − δ
                          2
     2n tasks of task time δ




The optimal packing packs all tasks of task time 1 C + δ and 1 C − δ in n
                                                    2           2
stations, and all tasks δ in 1 station. Thus giving n + 1 stations.
After the RPW sorting, the Next Fit algorithm packs into one station tasks
1                                    1
2 C + δ and δ and in another station 2 C − δ and δ. Thus giving 2n stations.
 M. Vong (SNU)                      SALBP                  December 15th 2009   22 / 24
Worst case ratio found in the instances


  Name        n      c      m∗    mFF     mRPW    CPU FF     CPU RPW         CPU B
  Arcus 2     111    5755   27    29      27       < 1ms        < 1ms        0.016s
  Barthold2   148    85     50    55      53       0.016s       0.031s        3600s
  Heskiaoff    28     256    4     5       4        < 1ms        < 1ms        < 1ms
  Lutz2       89     14     37    43      40       < 1ms        < 1ms        3.094s
  Scholl      297    1699   42    43      42       0.015s       0.078s        2603s
  Warnecke    58     56     29    34      33       0.016s       0.015s       0.906s


    The previously devised algorithms were implemented using C programming
    language and tested on 20 instances provided by the literature.
    The optimal solution was obtained using an efficient Branch and Bound Algorithm
    (SALOME devised by Scholl).
    The RPW sorting as a preprocessing is found to ameliorate the objective value of
    the solution.



     M. Vong (SNU)                      SALBP                  December 15th 2009   23 / 24
Conclusion



   A simple case of Assembly Line Balancing Problem was studied.
   A simple heuristic was proposed and implemented using C
   programming language.
   Computational results were compared to optimal one obtained by the
   current best Branch and Bound algorithm.
   Worst case analysis of our simple heuristic was studied.
        We found out that the factor 2 is a tight worst case ratio even for the
        adaptation of First Fit algorithm.




    M. Vong (SNU)                   SALBP                  December 15th 2009   24 / 24

Más contenido relacionado

La actualidad más candente

Multilayer Neural Networks
Multilayer Neural NetworksMultilayer Neural Networks
Multilayer Neural NetworksESCOM
 
Distributed solution of stochastic optimal control problem on GPUs
Distributed solution of stochastic optimal control problem on GPUsDistributed solution of stochastic optimal control problem on GPUs
Distributed solution of stochastic optimal control problem on GPUsPantelis Sopasakis
 
Aaex7 group2(中英夾雜)
Aaex7 group2(中英夾雜)Aaex7 group2(中英夾雜)
Aaex7 group2(中英夾雜)Shiang-Yun Yang
 
Pseudo Random Number Generators
Pseudo Random Number GeneratorsPseudo Random Number Generators
Pseudo Random Number GeneratorsDarshini Parikh
 
Wk 12 fr bode plot nyquist may 9 2016
Wk 12 fr bode plot nyquist   may 9 2016Wk 12 fr bode plot nyquist   may 9 2016
Wk 12 fr bode plot nyquist may 9 2016Charlton Inao
 
Aaex4 group2(中英夾雜)
Aaex4 group2(中英夾雜)Aaex4 group2(中英夾雜)
Aaex4 group2(中英夾雜)Shiang-Yun Yang
 
Ultrasound Modular Architecture
Ultrasound Modular ArchitectureUltrasound Modular Architecture
Ultrasound Modular ArchitectureJose Miguel Moreno
 
Dcs lec02 - z-transform
Dcs   lec02 - z-transformDcs   lec02 - z-transform
Dcs lec02 - z-transformAmr E. Mohamed
 
Parallel Evaluation of Multi-Semi-Joins
Parallel Evaluation of Multi-Semi-JoinsParallel Evaluation of Multi-Semi-Joins
Parallel Evaluation of Multi-Semi-JoinsJonny Daenen
 
Bayesian Inference and Uncertainty Quantification for Inverse Problems
Bayesian Inference and Uncertainty Quantification for Inverse ProblemsBayesian Inference and Uncertainty Quantification for Inverse Problems
Bayesian Inference and Uncertainty Quantification for Inverse ProblemsMatt Moores
 
Time and space complexity
Time and space complexityTime and space complexity
Time and space complexityAnkit Katiyar
 
Digital Signal Processing[ECEG-3171]-Ch1_L03
Digital Signal Processing[ECEG-3171]-Ch1_L03Digital Signal Processing[ECEG-3171]-Ch1_L03
Digital Signal Processing[ECEG-3171]-Ch1_L03Rediet Moges
 

La actualidad más candente (20)

Bode lect
Bode lectBode lect
Bode lect
 
Multilayer Neural Networks
Multilayer Neural NetworksMultilayer Neural Networks
Multilayer Neural Networks
 
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
 
Recursive Compressed Sensing
Recursive Compressed SensingRecursive Compressed Sensing
Recursive Compressed Sensing
 
Distributed solution of stochastic optimal control problem on GPUs
Distributed solution of stochastic optimal control problem on GPUsDistributed solution of stochastic optimal control problem on GPUs
Distributed solution of stochastic optimal control problem on GPUs
 
Aaex7 group2(中英夾雜)
Aaex7 group2(中英夾雜)Aaex7 group2(中英夾雜)
Aaex7 group2(中英夾雜)
 
Pseudo Random Number Generators
Pseudo Random Number GeneratorsPseudo Random Number Generators
Pseudo Random Number Generators
 
Wk 12 fr bode plot nyquist may 9 2016
Wk 12 fr bode plot nyquist   may 9 2016Wk 12 fr bode plot nyquist   may 9 2016
Wk 12 fr bode plot nyquist may 9 2016
 
Aaex4 group2(中英夾雜)
Aaex4 group2(中英夾雜)Aaex4 group2(中英夾雜)
Aaex4 group2(中英夾雜)
 
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
 
Mlab i
Mlab iMlab i
Mlab i
 
QMC Program: Trends and Advances in Monte Carlo Sampling Algorithms Workshop,...
QMC Program: Trends and Advances in Monte Carlo Sampling Algorithms Workshop,...QMC Program: Trends and Advances in Monte Carlo Sampling Algorithms Workshop,...
QMC Program: Trends and Advances in Monte Carlo Sampling Algorithms Workshop,...
 
QMC Opening Workshop, Support Points - a new way to compact distributions, wi...
QMC Opening Workshop, Support Points - a new way to compact distributions, wi...QMC Opening Workshop, Support Points - a new way to compact distributions, wi...
QMC Opening Workshop, Support Points - a new way to compact distributions, wi...
 
PCA on graph/network
PCA on graph/networkPCA on graph/network
PCA on graph/network
 
Ultrasound Modular Architecture
Ultrasound Modular ArchitectureUltrasound Modular Architecture
Ultrasound Modular Architecture
 
Dcs lec02 - z-transform
Dcs   lec02 - z-transformDcs   lec02 - z-transform
Dcs lec02 - z-transform
 
Parallel Evaluation of Multi-Semi-Joins
Parallel Evaluation of Multi-Semi-JoinsParallel Evaluation of Multi-Semi-Joins
Parallel Evaluation of Multi-Semi-Joins
 
Bayesian Inference and Uncertainty Quantification for Inverse Problems
Bayesian Inference and Uncertainty Quantification for Inverse ProblemsBayesian Inference and Uncertainty Quantification for Inverse Problems
Bayesian Inference and Uncertainty Quantification for Inverse Problems
 
Time and space complexity
Time and space complexityTime and space complexity
Time and space complexity
 
Digital Signal Processing[ECEG-3171]-Ch1_L03
Digital Signal Processing[ECEG-3171]-Ch1_L03Digital Signal Processing[ECEG-3171]-Ch1_L03
Digital Signal Processing[ECEG-3171]-Ch1_L03
 

Destacado

Assembly line balancing
Assembly line balancingAssembly line balancing
Assembly line balancingRohit Goutam
 
Sergio Antonio Salvi, Technical Representation Plan (lecture extract)
Sergio Antonio Salvi, Technical Representation Plan (lecture extract)Sergio Antonio Salvi, Technical Representation Plan (lecture extract)
Sergio Antonio Salvi, Technical Representation Plan (lecture extract)Sergio Antonio Salvi
 
Sergio Antonio Salvi, Design for Assembly (lecture extract)
Sergio Antonio Salvi, Design for Assembly (lecture extract)Sergio Antonio Salvi, Design for Assembly (lecture extract)
Sergio Antonio Salvi, Design for Assembly (lecture extract)Sergio Antonio Salvi
 
Memory management early_systems
Memory management early_systemsMemory management early_systems
Memory management early_systemsMybej Che
 
Design for manufacturing & assembly report (table lamp)
Design for manufacturing & assembly report (table lamp)Design for manufacturing & assembly report (table lamp)
Design for manufacturing & assembly report (table lamp)hrishik26
 
Design for Manufacturing and Assembly
Design for Manufacturing and AssemblyDesign for Manufacturing and Assembly
Design for Manufacturing and Assemblyelai1001
 
Assembly Line Balancing -Example
Assembly Line Balancing -ExampleAssembly Line Balancing -Example
Assembly Line Balancing -ExampleJoseph Konnully
 
Design for Assembly (DFA)
Design for Assembly (DFA)Design for Assembly (DFA)
Design for Assembly (DFA)Richard Farr
 

Destacado (13)

Assembly line balancing
Assembly line balancingAssembly line balancing
Assembly line balancing
 
Sergio Antonio Salvi, Technical Representation Plan (lecture extract)
Sergio Antonio Salvi, Technical Representation Plan (lecture extract)Sergio Antonio Salvi, Technical Representation Plan (lecture extract)
Sergio Antonio Salvi, Technical Representation Plan (lecture extract)
 
Sergio Antonio Salvi, Design for Assembly (lecture extract)
Sergio Antonio Salvi, Design for Assembly (lecture extract)Sergio Antonio Salvi, Design for Assembly (lecture extract)
Sergio Antonio Salvi, Design for Assembly (lecture extract)
 
Memory management early_systems
Memory management early_systemsMemory management early_systems
Memory management early_systems
 
Design For Assembly
Design For AssemblyDesign For Assembly
Design For Assembly
 
Design for manufacturing & assembly report (table lamp)
Design for manufacturing & assembly report (table lamp)Design for manufacturing & assembly report (table lamp)
Design for manufacturing & assembly report (table lamp)
 
Dfma
DfmaDfma
Dfma
 
Design for Manufacturing and Assembly
Design for Manufacturing and AssemblyDesign for Manufacturing and Assembly
Design for Manufacturing and Assembly
 
Assembly Line Balancing -Example
Assembly Line Balancing -ExampleAssembly Line Balancing -Example
Assembly Line Balancing -Example
 
Design for Assembly (DFA)
Design for Assembly (DFA)Design for Assembly (DFA)
Design for Assembly (DFA)
 
DFMA design for manufacturing and assembly
DFMA design for manufacturing and assembly DFMA design for manufacturing and assembly
DFMA design for manufacturing and assembly
 
assembly line balancing
assembly line balancingassembly line balancing
assembly line balancing
 
Build Features, Not Apps
Build Features, Not AppsBuild Features, Not Apps
Build Features, Not Apps
 

Similar a A study of the worst case ratio of a simple algorithm for simple assembly line balancing problem

lecture01_lecture01_lecture0001_ceva.pdf
lecture01_lecture01_lecture0001_ceva.pdflecture01_lecture01_lecture0001_ceva.pdf
lecture01_lecture01_lecture0001_ceva.pdfAnaNeacsu5
 
Stochastic Alternating Direction Method of Multipliers
Stochastic Alternating Direction Method of MultipliersStochastic Alternating Direction Method of Multipliers
Stochastic Alternating Direction Method of MultipliersTaiji Suzuki
 
[AAAI2021] Combinatorial Pure Exploration with Full-bandit or Partial Linear ...
[AAAI2021] Combinatorial Pure Exploration with Full-bandit or Partial Linear ...[AAAI2021] Combinatorial Pure Exploration with Full-bandit or Partial Linear ...
[AAAI2021] Combinatorial Pure Exploration with Full-bandit or Partial Linear ...Yuko Kuroki (黒木祐子)
 
Low-rank response surface in numerical aerodynamics
Low-rank response surface in numerical aerodynamicsLow-rank response surface in numerical aerodynamics
Low-rank response surface in numerical aerodynamicsAlexander Litvinenko
 
5.3 dynamic programming 03
5.3 dynamic programming 035.3 dynamic programming 03
5.3 dynamic programming 03Krish_ver2
 
Optimization Techniques
Optimization TechniquesOptimization Techniques
Optimization TechniquesAjay Bidyarthy
 
ACM ICPC 2015 NEERC (Northeastern European Regional Contest) Problems Review
ACM ICPC 2015 NEERC (Northeastern European Regional Contest) Problems ReviewACM ICPC 2015 NEERC (Northeastern European Regional Contest) Problems Review
ACM ICPC 2015 NEERC (Northeastern European Regional Contest) Problems ReviewRoman Elizarov
 
Estimating Future Initial Margin with Machine Learning
Estimating Future Initial Margin with Machine LearningEstimating Future Initial Margin with Machine Learning
Estimating Future Initial Margin with Machine LearningAndres Hernandez
 
EENG519FinalProjectReport
EENG519FinalProjectReportEENG519FinalProjectReport
EENG519FinalProjectReportDaniel K
 
Sloshing-aware MPC for upper stage attitude control
Sloshing-aware MPC for upper stage attitude controlSloshing-aware MPC for upper stage attitude control
Sloshing-aware MPC for upper stage attitude controlPantelis Sopasakis
 
UNIT-II.pptx
UNIT-II.pptxUNIT-II.pptx
UNIT-II.pptxJyoReddy9
 
MVPA with SpaceNet: sparse structured priors
MVPA with SpaceNet: sparse structured priorsMVPA with SpaceNet: sparse structured priors
MVPA with SpaceNet: sparse structured priorsElvis DOHMATOB
 
Optimal L-shaped matrix reordering, aka graph's core-periphery
Optimal L-shaped matrix reordering, aka graph's core-peripheryOptimal L-shaped matrix reordering, aka graph's core-periphery
Optimal L-shaped matrix reordering, aka graph's core-peripheryFrancesco Tudisco
 
Random Matrix Theory and Machine Learning - Part 3
Random Matrix Theory and Machine Learning - Part 3Random Matrix Theory and Machine Learning - Part 3
Random Matrix Theory and Machine Learning - Part 3Fabian Pedregosa
 
Lec04 min cost linear problems
Lec04 min cost linear problemsLec04 min cost linear problems
Lec04 min cost linear problemsDanny Luk
 

Similar a A study of the worst case ratio of a simple algorithm for simple assembly line balancing problem (20)

QCD Phase Diagram
QCD Phase DiagramQCD Phase Diagram
QCD Phase Diagram
 
lecture01_lecture01_lecture0001_ceva.pdf
lecture01_lecture01_lecture0001_ceva.pdflecture01_lecture01_lecture0001_ceva.pdf
lecture01_lecture01_lecture0001_ceva.pdf
 
Stochastic Alternating Direction Method of Multipliers
Stochastic Alternating Direction Method of MultipliersStochastic Alternating Direction Method of Multipliers
Stochastic Alternating Direction Method of Multipliers
 
QMC: Transition Workshop - Applying Quasi-Monte Carlo Methods to a Stochastic...
QMC: Transition Workshop - Applying Quasi-Monte Carlo Methods to a Stochastic...QMC: Transition Workshop - Applying Quasi-Monte Carlo Methods to a Stochastic...
QMC: Transition Workshop - Applying Quasi-Monte Carlo Methods to a Stochastic...
 
[AAAI2021] Combinatorial Pure Exploration with Full-bandit or Partial Linear ...
[AAAI2021] Combinatorial Pure Exploration with Full-bandit or Partial Linear ...[AAAI2021] Combinatorial Pure Exploration with Full-bandit or Partial Linear ...
[AAAI2021] Combinatorial Pure Exploration with Full-bandit or Partial Linear ...
 
Talk iccf 19_ben_hammouda
Talk iccf 19_ben_hammoudaTalk iccf 19_ben_hammouda
Talk iccf 19_ben_hammouda
 
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
Program on Quasi-Monte Carlo and High-Dimensional Sampling Methods for Applie...
 
Low-rank response surface in numerical aerodynamics
Low-rank response surface in numerical aerodynamicsLow-rank response surface in numerical aerodynamics
Low-rank response surface in numerical aerodynamics
 
5.3 dynamic programming 03
5.3 dynamic programming 035.3 dynamic programming 03
5.3 dynamic programming 03
 
Optimization Techniques
Optimization TechniquesOptimization Techniques
Optimization Techniques
 
ACM ICPC 2015 NEERC (Northeastern European Regional Contest) Problems Review
ACM ICPC 2015 NEERC (Northeastern European Regional Contest) Problems ReviewACM ICPC 2015 NEERC (Northeastern European Regional Contest) Problems Review
ACM ICPC 2015 NEERC (Northeastern European Regional Contest) Problems Review
 
Estimating Future Initial Margin with Machine Learning
Estimating Future Initial Margin with Machine LearningEstimating Future Initial Margin with Machine Learning
Estimating Future Initial Margin with Machine Learning
 
EENG519FinalProjectReport
EENG519FinalProjectReportEENG519FinalProjectReport
EENG519FinalProjectReport
 
Sloshing-aware MPC for upper stage attitude control
Sloshing-aware MPC for upper stage attitude controlSloshing-aware MPC for upper stage attitude control
Sloshing-aware MPC for upper stage attitude control
 
UNIT-II.pptx
UNIT-II.pptxUNIT-II.pptx
UNIT-II.pptx
 
MVPA with SpaceNet: sparse structured priors
MVPA with SpaceNet: sparse structured priorsMVPA with SpaceNet: sparse structured priors
MVPA with SpaceNet: sparse structured priors
 
Subquad multi ff
Subquad multi ffSubquad multi ff
Subquad multi ff
 
Optimal L-shaped matrix reordering, aka graph's core-periphery
Optimal L-shaped matrix reordering, aka graph's core-peripheryOptimal L-shaped matrix reordering, aka graph's core-periphery
Optimal L-shaped matrix reordering, aka graph's core-periphery
 
Random Matrix Theory and Machine Learning - Part 3
Random Matrix Theory and Machine Learning - Part 3Random Matrix Theory and Machine Learning - Part 3
Random Matrix Theory and Machine Learning - Part 3
 
Lec04 min cost linear problems
Lec04 min cost linear problemsLec04 min cost linear problems
Lec04 min cost linear problems
 

A study of the worst case ratio of a simple algorithm for simple assembly line balancing problem

  • 1. A study on a simple heuristic for Simple Assembly Line Balancing problem Monard VONG Optimization Lab. Seoul National University December 15th 2009 M. Vong (SNU) SALBP December 15th 2009 1 / 24
  • 2. Outline 1 Introduction to Assembly Line Problems 2 Solving SALBP 3 Worst case performance ratio and computational experiment M. Vong (SNU) SALBP December 15th 2009 2 / 24
  • 3. Concept of Assembly Line (AL) Figure: Tasks and precedence constraints Figure: Assembly Line Concept In assembly line balancing problem, the process (tasks The amount of time a and constraints) is known workpiece can be processed by a station, before the conveyor The goal is to find a feasible belt moves the workpiece to assignment of tasks to the next station is called the stations. cycle time. M. Vong (SNU) SALBP December 15th 2009 3 / 24
  • 4. Simple Assembly Line Balancing Problem Now we consider the simplest case of an AL problem: Simple Assembly Line Balancing Problem The objective is to maximize the efficiency of the line. Characteristics 1 paced serial line; 2 deterministic (and integral) task time tj ; 3 assignment restrictions: precedence constraints ; From now on, we focus on SALBP-1: cycle time c is fixed minimize the number of stations required This problem is NP-hard. M. Vong (SNU) SALBP December 15th 2009 4 / 24
  • 5. SALBP-1 as an optimization problem Input Set of task V = {1, . . . , n} with task time tj ∈ N, ∀j ∈ V Precedence digraph G = (V , A) (partial ordering of tasks) Cycle time c Goal Find number of station m to maximize the line efficiency. max tsum m·c with total task time tsum = n tj j=1 Output S1 , . . . , Sm , m subset of tasks such as ∪k∈1,...,m Sk = V and Sk is feasible for all station. M. Vong (SNU) SALBP December 15th 2009 5 / 24
  • 6. Notations n Number of tasks; V Set of task V = 1, . . . , n m Number of stations; index k = 1, . . . , m m∗ Optimal number of station m, m Lower, upper bound on m∗ tj Task time of task j = 1, . . . , n Pj Set of direct predecessors of task j Pj∗ Set of all predecessors of task j Fj∗ Set of all followers of task j Sk Set of task assigned to station k: Station load t(Sk ) t(Sk ) = j∈Sk tj , k = 1, . . . , m M. Vong (SNU) SALBP December 15th 2009 6 / 24
  • 7. State of the Art Lot of research done on SALBP since 50 years. Many heuristics and optimal procedures have been designed. Most research is focused on solving the problem with additional constraints. But there is few research on worst case ratio of heuristic algorithms. Queyranne(1985): no polynomial time algorithm achieves an absolute worst case performance ratio less than 3 2 M. Vong (SNU) SALBP December 15th 2009 7 / 24
  • 8. Outline 1 Introduction to Assembly Line Problems 2 Solving SALBP 3 Worst case performance ratio and computational experiment M. Vong (SNU) SALBP December 15th 2009 8 / 24
  • 9. Construction schemes for heuristic algorithm Most existing algorithm enumerates solutions by constructing them successively assigning tasks or subset of tasks to stations. Definition Availability A task j is available if all predecessor h ∈ Pj∗ have been assigned. Assignability An available task j is assignable to a station k if the current idle time of k is sufficient. Maximal Station Load A station load Sk is maximal if no available task is assignable to k. M. Vong (SNU) SALBP December 15th 2009 9 / 24
  • 10. Task-Oriented Greedy Heuristic for SALBP-1 Problem Adaptation of Next-Fit algorithm 1: topologically sort the tasks 2: k = 1 3: for i = 1 → n do 4: if t(Sk ) + ti ≤ c then 5: Sk ← i 6: else 7: k ←k +1 8: end if 9: end for M. Vong (SNU) SALBP December 15th 2009 10 / 24
  • 11. For instance c = 10 Figure: Precedence Graph Figure: Solution Obtained by Next-Fit M. Vong (SNU) SALBP December 15th 2009 11 / 24
  • 12. Adaptation of First Fit algorithm 1: topologically sort the tasks 2: m ← 1 3: for i = 1 → n do 4: if ∃k ∈ {1, . . . , m} such that tSk + tj ≤ c and the precedence constraints are respected then 5: Sk ← Sk + {i} 6: else 7: m ←m+1 8: end if 9: end for M. Vong (SNU) SALBP December 15th 2009 12 / 24
  • 13. For instance c = 10 Figure: Precedence Graph Figure: Solution Obtained by adaptation of First-Fit M. Vong (SNU) SALBP December 15th 2009 13 / 24
  • 14. c = 10 Figure: An optimal Solution with 6 stations M. Vong (SNU) SALBP December 15th 2009 14 / 24
  • 15. Outline 1 Introduction to Assembly Line Problems 2 Solving SALBP 3 Worst case performance ratio and computational experiment M. Vong (SNU) SALBP December 15th 2009 15 / 24
  • 16. Worst case performance ratio Notations I is an instance of our optimization problem. z ∗ (I ) is the optimal value solution. z H (I ) is the value of the solution obtained by heuristic H The worst case ratio of H is defined as the smallest constant ρH such that z H (I ) ≤ ρH z ∗ (I ) for each instance of I ∈ I The asymptotic worst case ratio of H is defined as the smallest constant ρH for which there exist another constant δ such that z H (I ) ≤ ρH z ∗ (I ) + δ for each instance of I ∈ I. M. Vong (SNU) SALBP December 15th 2009 16 / 24
  • 17. Upper Bound on the worst case ratio Proposition An upper bound of the worst case ratio of any heuristics which build maximal station load is 2. Proof. For any heuristic H which build maximal station load, for any station k built by this heuristic: t(Sk ) + t(Sk+1 ) > c mH n mH mH ·c t(Sk ) j=1 tj k=1t(Sk ) > 2 . However, m∗ ≥ k=1 c = c , then we obtain m < 2 · m∗ . H Corollary The worst case ratio of Next Fit is 2, and it is tight. M. Vong (SNU) SALBP December 15th 2009 17 / 24
  • 18. A 2 worst case instance for SALBP We will provide an instance for which, even the adaptation of First Fit algorithm to SALBP cannot beat the 2 worst case factor. 1C Let consider 0 < δ ≤ 6 k: 2k tasks of task time 3δ, 2k tasks of task time 1 C − 2δ, 2 2 ∗ (2k − 1) tasks of task time δ Figure: 2 Worst case instance M. Vong (SNU) SALBP December 15th 2009 18 / 24
  • 19. Figure: Optimal Solution Figure: Solution given by the algorithm A(I ) 2k lim = lim =2 k→∞ OPT (I ) k→∞ k + 1 M. Vong (SNU) SALBP December 15th 2009 19 / 24
  • 20. Ranked Positional Weight Using a topological sorting is not enough, the quality of the sorting may be improved. Ranking Positional Weight (RPW) of task i is the sum of the task time ti and the task time of all followers of i. RPWi = ti + j∈F ∗ tj i Tasks are sorted by non-increasing RPW. Example RPW1 = 38 RPW6 = 18 RPW2 = 23 RPW7 = 17 RPW3 = 32 RPW8 = 13 RPW4 = 27 RPW9 = 11 RPW5 = 22 RPW10 = 2 Sorting by non increasing RPW, the following order is considered. 1, 3, 4, 2, 5, 6, 7, 8, 9, 10 M. Vong (SNU) SALBP December 15th 2009 20 / 24
  • 21. Using RPW as topological sorting, the optimal solution is obtained for the previous worst case instance with the same first fit algorithm. Ranked Positional Weight algorithm 1: Sort the tasks by non-increasing ranked positional weight 2: Use First Fit rule What about the worst case of the improved algorithm then? M. Vong (SNU) SALBP December 15th 2009 21 / 24
  • 22. If we used Next fit algorithm, instead of First Fit, it is still possible to find instances for which the (asymptotic) worst case ratio is 2. Consider n integer and δ > 0 such that 2nδ < 1. The instance is made of: n tasks of task time 1 C + δ 2 n tasks of task time 1 C − δ 2 2n tasks of task time δ The optimal packing packs all tasks of task time 1 C + δ and 1 C − δ in n 2 2 stations, and all tasks δ in 1 station. Thus giving n + 1 stations. After the RPW sorting, the Next Fit algorithm packs into one station tasks 1 1 2 C + δ and δ and in another station 2 C − δ and δ. Thus giving 2n stations. M. Vong (SNU) SALBP December 15th 2009 22 / 24
  • 23. Worst case ratio found in the instances Name n c m∗ mFF mRPW CPU FF CPU RPW CPU B Arcus 2 111 5755 27 29 27 < 1ms < 1ms 0.016s Barthold2 148 85 50 55 53 0.016s 0.031s 3600s Heskiaoff 28 256 4 5 4 < 1ms < 1ms < 1ms Lutz2 89 14 37 43 40 < 1ms < 1ms 3.094s Scholl 297 1699 42 43 42 0.015s 0.078s 2603s Warnecke 58 56 29 34 33 0.016s 0.015s 0.906s The previously devised algorithms were implemented using C programming language and tested on 20 instances provided by the literature. The optimal solution was obtained using an efficient Branch and Bound Algorithm (SALOME devised by Scholl). The RPW sorting as a preprocessing is found to ameliorate the objective value of the solution. M. Vong (SNU) SALBP December 15th 2009 23 / 24
  • 24. Conclusion A simple case of Assembly Line Balancing Problem was studied. A simple heuristic was proposed and implemented using C programming language. Computational results were compared to optimal one obtained by the current best Branch and Bound algorithm. Worst case analysis of our simple heuristic was studied. We found out that the factor 2 is a tight worst case ratio even for the adaptation of First Fit algorithm. M. Vong (SNU) SALBP December 15th 2009 24 / 24