SlideShare a Scribd company logo
A N I M P R O V E M E N T O N T H E A L D E P H E U R I S T I C
V I A D E P A R T M E N T B A T C H I N G
ALDEP++
BACKGROUND & MOTIVATION
Introduction to ALDEP –
• Construction algorithm that randomly chooses the first department
• Successively adds more on the basis of closeness ratings of the relationship chart following S
pattern
Advantages of ALDEP –
• simple and not time consuming.
• does not need any initial Layout
• Since the S-flow method is used in ALDEP, resulting in practical layouts
• A greater variety of good options are exposed as compared to other methods
BACKGROUND & MOTIVATION
Shortcomings of ALDEP -
• Shortsightedness - Only considers relationship between immediate previously placed department
and each of remaining departments.
• Each individual department is Considered for one subsequent department decision
• Leads to generation of many suboptimal layouts, not necessarily the most optimal layout.
• Blind to constraints (such as disallowing departments with X relation)
• Only considers adjacent departments for comparing different layouts via total closeness ratings,
there is information lost regarding the total costs.
Motivation –
• The ALDEP adds one department at a time resulting in blindness to many department
relationships
• Adding all departments at once makes it a quadratic assignment problem which is
computationally difficult as the problem size increases.
• ALDEP++ is a method which functions between these two extremes by adding departments in
small batches.
• ALDEP provides a good starting point for designing a new constructive heuristic algorithm.
• New algorithm “ALDEP++” will attempt to counter the shortcomings of ALDEP without
abandoning these advantages.
EVOLUTION OF ALDEP++
ALDEP++ : The General Approach
• Addition of departments in batches
• Use of flow * distance as our measurement for what order to place the departments from each batch
in the ALDEP++ algorithm.
• Slightly modified input compared to ALDEP.
• In-between flow matrix is required
Benefits of ALDEP++
• Addition of department in batches overcomes the shortsightedness of ALDEP and produce more
optimized solutions
• Use of flow * distance as a measure , it allows us to obtain total costs of comparable to layouts
produced with the MicroCRAFT algorithm while still obtaining high TCR’s (maintaining some of the
simplicity of the ALDEP algorithm)
MODEL FORMULATION :ALDEP++
Assumptions –
• The area of departments is a multiple of sweep width.
• The length and width of the building is a multiple of sweep width.
• The floor space should be amenable to the serpentine flow.
• If there is a tie, departments are chosen/placed randomly.
• The shape of department can be changed.
• The next department starts where the previous one ends.
Batching -
• Uses relationship categories with respective scoring weights to determine which departments to
add next.
• Takes the sum of the relationships to all departments from the previous batch.
• Ranks the remaining departments by this relationship sum, the top scoring departments can be
determined as the subset to be included in the next batch.
MODEL FORMULATION :ALDEP++
Order Determination
• Batch of n departments analyze all n! possible orders of placement along the serpentine sweep pattern
• Resulting combinations are compared using a flow*distance calculation for the subset of departments
consisting of the current batch and the previous batch
• The placement order resulting in the lowest total flow*distance is selected and applied
• Distances can be measured as either E-DIST or centroid distances, depending on the preferences and
limitations of the user
Initialization
• First n departments are to be placed following the ALDEP algorithm.
• The first department is chosen at random, and each successive department is chosen based on the highest
relationships with the previously placed department. This process introduces the randomness of the ALDEP
process.
Iterations
• ALDEP++ makes random decisions whenever there are ties.
• At the very least, the starting department will be random.
• If multiple layouts are tied in the end for highest TCR, total flow * distance is used as the tiebreaker.
EXAMPLE MODEL: ALDEP++
Flow-Between Matrix
A B C D E F
A 0 8 0 4 0
B 0 5 0 2
C 0 1 0
D 6 0
E 4
F
Department Size (sq units)
A 400
B 800
C 600
D 600
E 800
F 400
Total Facility Size = 60 x 60 units
Sweep Width = 20 units
Relationship Categories
A = 8
E = 6, 5
I = 4
O = 1, 2
U = 0
Scoring Weights
A = 64
E = 16
I = 4
O = 1
U = 0
STEP 1 –
• Department F randomly selected for the start, and
the ALDEP algorithm selected E as the next
department as it has the strongest relationship with
F
• Using a batch size of 2
Initial two departments (first batch)
F
E
EXAMPLE MODEL: ALDEP++
Flow-Between Matrix
A B C D E F
A 0 8 0 4 0
B 0 5 0 2
C 0 1 0
D 6 0
E 4
F
Department Size (sq units)
A 400
B 800
C 600
D 600
E 800
F 400
STEP 2 – Second batch:
• Calculating the relationship ratings based on the scoring weights, it is determined that A
and D are the departments for the second batch.
• Because n=2, there are 2!=2 order combinations
Department Relationship Rating
A 4+0 = 4
B 0+1 = 1
C 1+0 = 1
D 16+0 = 16
F
A
E
D
F
D
E
A
Alternative 1 Alternative 2
EXAMPLE MODEL: ALDEP++
Flow-Between Matrix
A B C D E F
A 0 8 0 4 0
B 0 5 0 2
C 0 1 0
D 6 0
E 4
F
F
A
E
D
Alternative 1 Alternative 2
Calculating the sum-product of the E-DIST matrices with the flow matrix gives 490 for
order 1 and 364 for order 2. Because we want the minimum flow*distance, we select
alternative 2.
EXAMPLE MODEL: ALDEP++
Flow-Between Matrix
A B C D E F
A 0 8 0 4 0
B 0 5 0 2
C 0 1 0
D 6 0
E 4
F
F B
A
E
D C
Alternative 1 Alternative 2
• This is the last batch, so we can immediately skip to testing the 2 order combinations.
• Calculating the sum products gives 603.75 for order 1 and 371.68 for order 2. Therefore we
choose order 2, and that is the completed layout.
Third Batch:
F C
A
E
D B
Alternative 1 Alternative 2
EXAMPLE MODEL: ALDEP++
• This final layout has a total flow*distance of 1066.7 as compared to the results from the standard ALDEP
procedure which results in a higher total of 1132.9.
• Under the adjacency scoring method, the ALDEP++ layout receives a total weighted score of 104 as
compared to the ALDEP layout which scores only 101.
• In this example, ALDEP++ has outperformed ALDEP under both methods of comparison.
F B
A
E
D C
F C
A
E
D B
ALDEP LAYOUT
total weighted score – 101
Flow*dist = 1132.9
ALDEP++ LAYOUT
Total weighted score – 104
Flow*dist = 1132.9
IMPLEMENTATION & RESULTS
IMPLEMENTATION –
• ALDEP++ was compared with the original ALDEP and MicroCRAFT
• All three heuristic algorithms implanted with Matlab code
• ALDEP++ was coded using centroid distances primary reason for this decision being making a fair
comparison in total cost with the results from MicroCRAFT, which also is based on centroid
distance.
• Layout problems with appropriate variability generated and the run in these three codes
RESULTS :
For all cases, our outputs were TCR (total closeness rating) for adjacency based scoring, and
centroid distance * flows for the cost based scoring.
IMPLEMENTATION & RESULTS
Interpretation :
• TCR results for the ALDEP++ layouts are generally very close to the TCR results from ALDEP
• Comparing case-by-case, it can be seen that ALDEP++ works better when:
• The layout is square or close to square.
• Department sizes are not highly variable.
• Each batch roughly fills one column of the serpentine flow pattern.
Total Material Handling Cost Comparison
ALDEP++ yields lower costs on average because ALDEP does not consider flow * distance in it’s algorithm, while
ALDEP++ does.
ALDEP++ gives significantly better TCR than MicroCRAFT. As a result, while ALDEP++ consistently has high
TCR, MicroCRAFT results in highly variable TCR’s
BENEFITS & CONCLUSIONS
Benefits -
• ALDEP++ procedure improves on the performance of the layouts generated through adding departments
in batches.
• In general, we have found that the ALDEP++ algorithm is producing lower cost layouts when compared
to the total costs obtained from the ALDEP algorithm.
• The layout TCR’s obtained from the ALDEP++ algorithm are not significantly lower than the TCR’s seen
from the ALDEP algorithm.
• ALDEP++ has the best properties of both MicroCRAFT and ALDEP algorithms.
• ALDEP++ also has the advantage of covering one of ALDEP’s flaws, in avoiding undesirable X-
relationships.
• Batch-approach of ALDEP++ allows it to avoid X-relationships more effectively
Conclusions –
• ALDEP++ is a feasible heuristic algorithm for generating new layouts
• Provides balance between adjacency based TCR scoring and distance * flow total cost scoring
• ALDEP++ performs similarly to ALDEP with respect to TCR, and also obtains low-cost layouts similar to
the results from MicroCRAFT.
• Generally, ALDEP++ works best when the factors allow single batches to fit in one pass of the serpentine
flow pattern
• As total cost is an important realistic consideration for comparing layouts, ALDEP++ offers facility layouts
that are more practically optimal as compared to ALDEP.

More Related Content

What's hot

APT part programming
APT part programmingAPT part programming
APT part programming
nmahi96
 
CIM 15ME62 Module-2: CAD and Computer Graphics Software; Module 3: Line Bal...
CIM 15ME62   Module-2: CAD and Computer Graphics Software; Module 3: Line Bal...CIM 15ME62   Module-2: CAD and Computer Graphics Software; Module 3: Line Bal...
CIM 15ME62 Module-2: CAD and Computer Graphics Software; Module 3: Line Bal...
Dr. Bhimsen Soragaon
 
Cim lab manual (10 mel77) by mohammed imran
Cim lab manual (10 mel77)  by mohammed imranCim lab manual (10 mel77)  by mohammed imran
Cim lab manual (10 mel77) by mohammed imran
Mohammed Imran
 
Unit 1 notes
Unit 1 notesUnit 1 notes
Unit 1 notes
Mahesh Pancity
 
COMPUTER AIDED PROCESS PLANNING (CAPP)
COMPUTER AIDED PROCESS PLANNING (CAPP)COMPUTER AIDED PROCESS PLANNING (CAPP)
COMPUTER AIDED PROCESS PLANNING (CAPP)
Victor Al
 
Flexible manufacturing systems (FMS)
Flexible manufacturing systems (FMS)Flexible manufacturing systems (FMS)
Flexible manufacturing systems (FMS)
jntuhcej
 
COMPUTER AIDED PROCESS PLANNING
COMPUTER AIDED PROCESS PLANNINGCOMPUTER AIDED PROCESS PLANNING
COMPUTER AIDED PROCESS PLANNING
jntuhcej
 
Group technology
Group technologyGroup technology
Group technology
jntuhcej
 
travel chart for plant layout
travel chart for plant layouttravel chart for plant layout
travel chart for plant layout
Yash Patel
 
Group technology, Production flow Analysis
Group technology, Production flow AnalysisGroup technology, Production flow Analysis
Group technology, Production flow Analysis
KEERTHIVASAN RAVI
 
DNC machines tools
DNC machines toolsDNC machines tools
DNC machines tools
jntuhcej
 
Computer Integrated Manufacturing System
Computer Integrated Manufacturing SystemComputer Integrated Manufacturing System
Computer Integrated Manufacturing System
Saif Eye
 
Method study
Method studyMethod study
Method study
Md Jawed Akbar
 
Cmm
CmmCmm
Module 3 Space Determination and Area Allocation
Module 3 Space Determination and Area AllocationModule 3 Space Determination and Area Allocation
Module 3 Space Determination and Area Allocation
Rashmi Srinivas
 
MRP-II
MRP-IIMRP-II
Flexible Manufacturing System
Flexible Manufacturing SystemFlexible Manufacturing System
Flexible Manufacturing System
Desai Vaishali
 
AUTOMATED STORAGE AND RETRIEVAL SYSTEM
AUTOMATED STORAGE AND RETRIEVAL SYSTEMAUTOMATED STORAGE AND RETRIEVAL SYSTEM
AUTOMATED STORAGE AND RETRIEVAL SYSTEM
Mihir Patel
 
Industrial Facility Design
Industrial Facility DesignIndustrial Facility Design
Industrial Facility Design
Umar Yazdani
 

What's hot (20)

APT part programming
APT part programmingAPT part programming
APT part programming
 
CIM 15ME62 Module-2: CAD and Computer Graphics Software; Module 3: Line Bal...
CIM 15ME62   Module-2: CAD and Computer Graphics Software; Module 3: Line Bal...CIM 15ME62   Module-2: CAD and Computer Graphics Software; Module 3: Line Bal...
CIM 15ME62 Module-2: CAD and Computer Graphics Software; Module 3: Line Bal...
 
Cim lab manual (10 mel77) by mohammed imran
Cim lab manual (10 mel77)  by mohammed imranCim lab manual (10 mel77)  by mohammed imran
Cim lab manual (10 mel77) by mohammed imran
 
Unit 1 notes
Unit 1 notesUnit 1 notes
Unit 1 notes
 
Facility Layout
Facility LayoutFacility Layout
Facility Layout
 
COMPUTER AIDED PROCESS PLANNING (CAPP)
COMPUTER AIDED PROCESS PLANNING (CAPP)COMPUTER AIDED PROCESS PLANNING (CAPP)
COMPUTER AIDED PROCESS PLANNING (CAPP)
 
Flexible manufacturing systems (FMS)
Flexible manufacturing systems (FMS)Flexible manufacturing systems (FMS)
Flexible manufacturing systems (FMS)
 
COMPUTER AIDED PROCESS PLANNING
COMPUTER AIDED PROCESS PLANNINGCOMPUTER AIDED PROCESS PLANNING
COMPUTER AIDED PROCESS PLANNING
 
Group technology
Group technologyGroup technology
Group technology
 
travel chart for plant layout
travel chart for plant layouttravel chart for plant layout
travel chart for plant layout
 
Group technology, Production flow Analysis
Group technology, Production flow AnalysisGroup technology, Production flow Analysis
Group technology, Production flow Analysis
 
DNC machines tools
DNC machines toolsDNC machines tools
DNC machines tools
 
Computer Integrated Manufacturing System
Computer Integrated Manufacturing SystemComputer Integrated Manufacturing System
Computer Integrated Manufacturing System
 
Method study
Method studyMethod study
Method study
 
Cmm
CmmCmm
Cmm
 
Module 3 Space Determination and Area Allocation
Module 3 Space Determination and Area AllocationModule 3 Space Determination and Area Allocation
Module 3 Space Determination and Area Allocation
 
MRP-II
MRP-IIMRP-II
MRP-II
 
Flexible Manufacturing System
Flexible Manufacturing SystemFlexible Manufacturing System
Flexible Manufacturing System
 
AUTOMATED STORAGE AND RETRIEVAL SYSTEM
AUTOMATED STORAGE AND RETRIEVAL SYSTEMAUTOMATED STORAGE AND RETRIEVAL SYSTEM
AUTOMATED STORAGE AND RETRIEVAL SYSTEM
 
Industrial Facility Design
Industrial Facility DesignIndustrial Facility Design
Industrial Facility Design
 

Viewers also liked

Close Loop MRP
Close Loop MRPClose Loop MRP
Close Loop MRP
guest865c0e0c
 
1.introduction of production and operations management
1.introduction of production and operations management1.introduction of production and operations management
1.introduction of production and operations managementAkash Bakshi
 
Analyzed operations of Groundwork Buffalo to identify Non-Value Add activitie...
Analyzed operations of Groundwork Buffalo to identify Non-Value Add activitie...Analyzed operations of Groundwork Buffalo to identify Non-Value Add activitie...
Analyzed operations of Groundwork Buffalo to identify Non-Value Add activitie...
Palaque Thakur
 
Layout10
Layout10Layout10
Layout10
Mukesh Verma
 
Plant Layout
Plant LayoutPlant Layout
Plant Layout
learnito
 
Adidas Brand Study
Adidas Brand StudyAdidas Brand Study
Adidas Brand Study
Madhusudan Partani
 
ABC
ABCABC
Inventory Control ABC Analysis
Inventory Control ABC AnalysisInventory Control ABC Analysis
Inventory Control ABC Analysisyashpal01
 
Forecasting Techniques
Forecasting TechniquesForecasting Techniques
Forecasting Techniques
guest865c0e0c
 

Viewers also liked (13)

Mac
MacMac
Mac
 
Close Loop MRP
Close Loop MRPClose Loop MRP
Close Loop MRP
 
production management
production managementproduction management
production management
 
1.introduction of production and operations management
1.introduction of production and operations management1.introduction of production and operations management
1.introduction of production and operations management
 
Planning (6) 2
Planning (6) 2Planning (6) 2
Planning (6) 2
 
Analyzed operations of Groundwork Buffalo to identify Non-Value Add activitie...
Analyzed operations of Groundwork Buffalo to identify Non-Value Add activitie...Analyzed operations of Groundwork Buffalo to identify Non-Value Add activitie...
Analyzed operations of Groundwork Buffalo to identify Non-Value Add activitie...
 
Layout10
Layout10Layout10
Layout10
 
Plant Layout
Plant LayoutPlant Layout
Plant Layout
 
Adidas Brand Study
Adidas Brand StudyAdidas Brand Study
Adidas Brand Study
 
ABC
ABCABC
ABC
 
Inventory Control ABC Analysis
Inventory Control ABC AnalysisInventory Control ABC Analysis
Inventory Control ABC Analysis
 
Plant layout
Plant layoutPlant layout
Plant layout
 
Forecasting Techniques
Forecasting TechniquesForecasting Techniques
Forecasting Techniques
 

Similar to ALDEP++: An improvement on the ALDEP heuristic via department batching.

Slicing of Object-Oriented Programs
Slicing of Object-Oriented ProgramsSlicing of Object-Oriented Programs
Slicing of Object-Oriented Programs
Praveen Penumathsa
 
Rseminarp
RseminarpRseminarp
Slicing of Object-Oriented Programs
Slicing of Object-Oriented ProgramsSlicing of Object-Oriented Programs
Slicing of Object-Oriented Programs
Praveen Penumathsa
 
Using Derivation-Free Optimization Methods in the Hadoop Cluster with Terasort
Using Derivation-Free Optimization Methods in the Hadoop Cluster with TerasortUsing Derivation-Free Optimization Methods in the Hadoop Cluster with Terasort
Using Derivation-Free Optimization Methods in the Hadoop Cluster with Terasort
Anhanguera Educacional S/A
 
Algorithm manual
Algorithm manualAlgorithm manual
Algorithm manual
EfendyMasuli
 
Algorithm Flowchart Manual ALGORITHM FLOWCHART MANUAL For STUDENTS
Algorithm   Flowchart Manual ALGORITHM   FLOWCHART MANUAL For STUDENTSAlgorithm   Flowchart Manual ALGORITHM   FLOWCHART MANUAL For STUDENTS
Algorithm Flowchart Manual ALGORITHM FLOWCHART MANUAL For STUDENTS
Alicia Edwards
 
detail of flowchart and algorithm that are used in programmingpdf
detail of flowchart and algorithm that are used in programmingpdfdetail of flowchart and algorithm that are used in programmingpdf
detail of flowchart and algorithm that are used in programmingpdf
ssuserf86fba
 
Using Derivation-Free Optimization in the Hadoop Cluster with Terasort
Using Derivation-Free Optimization in the Hadoop Cluster  with TerasortUsing Derivation-Free Optimization in the Hadoop Cluster  with Terasort
Using Derivation-Free Optimization in the Hadoop Cluster with Terasort
Anhanguera Educacional S/A
 
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Universitat Politècnica de Catalunya
 
(Slides) Efficient Evaluation Methods of Elementary Functions Suitable for SI...
(Slides) Efficient Evaluation Methods of Elementary Functions Suitable for SI...(Slides) Efficient Evaluation Methods of Elementary Functions Suitable for SI...
(Slides) Efficient Evaluation Methods of Elementary Functions Suitable for SI...
Naoki Shibata
 
High Speed radix256 algorithm using parallel prefix adder
High Speed radix256 algorithm using parallel prefix adderHigh Speed radix256 algorithm using parallel prefix adder
High Speed radix256 algorithm using parallel prefix adder
IJMER
 
Mod.2.pptx
Mod.2.pptxMod.2.pptx
Mod.2.pptx
Kokilak27
 
Directed Optimization on Pareto Frontier
Directed Optimization on Pareto FrontierDirected Optimization on Pareto Frontier
Directed Optimization on Pareto Frontier
eArtius, Inc.
 
Algorithm and C code related to data structure
Algorithm and C code related to data structureAlgorithm and C code related to data structure
Algorithm and C code related to data structure
Self-Employed
 
Low power correlation for IEEE 802.16 OFDM synchronisation using FPGA
Low power correlation for IEEE 802.16 OFDM  synchronisation using FPGA Low power correlation for IEEE 802.16 OFDM  synchronisation using FPGA
Low power correlation for IEEE 802.16 OFDM synchronisation using FPGA
Brundha Sholaganga
 
Data envelopment analysis
Data envelopment analysisData envelopment analysis
Data envelopment analysis
Prashant Chouhan
 
Improving the solution time of TIMES by playing with CPLEX/Barrier
Improving the solution time of TIMES by playing with CPLEX/BarrierImproving the solution time of TIMES by playing with CPLEX/Barrier
Improving the solution time of TIMES by playing with CPLEX/Barrier
IEA-ETSAP
 
IRJET- The RTL Model of a Reconfigurable Pipelined MCM
IRJET- The RTL Model of a Reconfigurable Pipelined MCMIRJET- The RTL Model of a Reconfigurable Pipelined MCM
IRJET- The RTL Model of a Reconfigurable Pipelined MCM
IRJET Journal
 
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
Tahmid Abtahi
 

Similar to ALDEP++: An improvement on the ALDEP heuristic via department batching. (20)

Slicing of Object-Oriented Programs
Slicing of Object-Oriented ProgramsSlicing of Object-Oriented Programs
Slicing of Object-Oriented Programs
 
Rseminarp
RseminarpRseminarp
Rseminarp
 
Slicing of Object-Oriented Programs
Slicing of Object-Oriented ProgramsSlicing of Object-Oriented Programs
Slicing of Object-Oriented Programs
 
Using Derivation-Free Optimization Methods in the Hadoop Cluster with Terasort
Using Derivation-Free Optimization Methods in the Hadoop Cluster with TerasortUsing Derivation-Free Optimization Methods in the Hadoop Cluster with Terasort
Using Derivation-Free Optimization Methods in the Hadoop Cluster with Terasort
 
Algorithm manual
Algorithm manualAlgorithm manual
Algorithm manual
 
Algorithm Flowchart Manual ALGORITHM FLOWCHART MANUAL For STUDENTS
Algorithm   Flowchart Manual ALGORITHM   FLOWCHART MANUAL For STUDENTSAlgorithm   Flowchart Manual ALGORITHM   FLOWCHART MANUAL For STUDENTS
Algorithm Flowchart Manual ALGORITHM FLOWCHART MANUAL For STUDENTS
 
detail of flowchart and algorithm that are used in programmingpdf
detail of flowchart and algorithm that are used in programmingpdfdetail of flowchart and algorithm that are used in programmingpdf
detail of flowchart and algorithm that are used in programmingpdf
 
Using Derivation-Free Optimization in the Hadoop Cluster with Terasort
Using Derivation-Free Optimization in the Hadoop Cluster  with TerasortUsing Derivation-Free Optimization in the Hadoop Cluster  with Terasort
Using Derivation-Free Optimization in the Hadoop Cluster with Terasort
 
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
Training Deep Networks with Backprop (D1L4 Insight@DCU Machine Learning Works...
 
(Slides) Efficient Evaluation Methods of Elementary Functions Suitable for SI...
(Slides) Efficient Evaluation Methods of Elementary Functions Suitable for SI...(Slides) Efficient Evaluation Methods of Elementary Functions Suitable for SI...
(Slides) Efficient Evaluation Methods of Elementary Functions Suitable for SI...
 
High Speed radix256 algorithm using parallel prefix adder
High Speed radix256 algorithm using parallel prefix adderHigh Speed radix256 algorithm using parallel prefix adder
High Speed radix256 algorithm using parallel prefix adder
 
Mod.2.pptx
Mod.2.pptxMod.2.pptx
Mod.2.pptx
 
Directed Optimization on Pareto Frontier
Directed Optimization on Pareto FrontierDirected Optimization on Pareto Frontier
Directed Optimization on Pareto Frontier
 
Algorithm and C code related to data structure
Algorithm and C code related to data structureAlgorithm and C code related to data structure
Algorithm and C code related to data structure
 
Low power correlation for IEEE 802.16 OFDM synchronisation using FPGA
Low power correlation for IEEE 802.16 OFDM  synchronisation using FPGA Low power correlation for IEEE 802.16 OFDM  synchronisation using FPGA
Low power correlation for IEEE 802.16 OFDM synchronisation using FPGA
 
Data envelopment analysis
Data envelopment analysisData envelopment analysis
Data envelopment analysis
 
Improving the solution time of TIMES by playing with CPLEX/Barrier
Improving the solution time of TIMES by playing with CPLEX/BarrierImproving the solution time of TIMES by playing with CPLEX/Barrier
Improving the solution time of TIMES by playing with CPLEX/Barrier
 
IRJET- The RTL Model of a Reconfigurable Pipelined MCM
IRJET- The RTL Model of a Reconfigurable Pipelined MCMIRJET- The RTL Model of a Reconfigurable Pipelined MCM
IRJET- The RTL Model of a Reconfigurable Pipelined MCM
 
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
A Framework for Scene Recognition Using Convolutional Neural Network as Featu...
 
Algorithm.pptx
Algorithm.pptxAlgorithm.pptx
Algorithm.pptx
 

Recently uploaded

Student information management system project report ii.pdf
Student information management system project report ii.pdfStudent information management system project report ii.pdf
Student information management system project report ii.pdf
Kamal Acharya
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
zwunae
 
Nuclear Power Economics and Structuring 2024
Nuclear Power Economics and Structuring 2024Nuclear Power Economics and Structuring 2024
Nuclear Power Economics and Structuring 2024
Massimo Talia
 
Immunizing Image Classifiers Against Localized Adversary Attacks
Immunizing Image Classifiers Against Localized Adversary AttacksImmunizing Image Classifiers Against Localized Adversary Attacks
Immunizing Image Classifiers Against Localized Adversary Attacks
gerogepatton
 
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
ydteq
 
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
bakpo1
 
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdfTop 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Teleport Manpower Consultant
 
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...
Amil Baba Dawood bangali
 
weather web application report.pdf
weather web application report.pdfweather web application report.pdf
weather web application report.pdf
Pratik Pawar
 
ethical hacking in wireless-hacking1.ppt
ethical hacking in wireless-hacking1.pptethical hacking in wireless-hacking1.ppt
ethical hacking in wireless-hacking1.ppt
Jayaprasanna4
 
ethical hacking-mobile hacking methods.ppt
ethical hacking-mobile hacking methods.pptethical hacking-mobile hacking methods.ppt
ethical hacking-mobile hacking methods.ppt
Jayaprasanna4
 
J.Yang, ICLR 2024, MLILAB, KAIST AI.pdf
J.Yang,  ICLR 2024, MLILAB, KAIST AI.pdfJ.Yang,  ICLR 2024, MLILAB, KAIST AI.pdf
J.Yang, ICLR 2024, MLILAB, KAIST AI.pdf
MLILAB
 
H.Seo, ICLR 2024, MLILAB, KAIST AI.pdf
H.Seo,  ICLR 2024, MLILAB,  KAIST AI.pdfH.Seo,  ICLR 2024, MLILAB,  KAIST AI.pdf
H.Seo, ICLR 2024, MLILAB, KAIST AI.pdf
MLILAB
 
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdfAKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
SamSarthak3
 
Gen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdfGen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdf
gdsczhcet
 
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
MdTanvirMahtab2
 
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
thanhdowork
 
Fundamentals of Electric Drives and its applications.pptx
Fundamentals of Electric Drives and its applications.pptxFundamentals of Electric Drives and its applications.pptx
Fundamentals of Electric Drives and its applications.pptx
manasideore6
 
Planning Of Procurement o different goods and services
Planning Of Procurement o different goods and servicesPlanning Of Procurement o different goods and services
Planning Of Procurement o different goods and services
JoytuBarua2
 
HYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generationHYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generation
Robbie Edward Sayers
 

Recently uploaded (20)

Student information management system project report ii.pdf
Student information management system project report ii.pdfStudent information management system project report ii.pdf
Student information management system project report ii.pdf
 
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
一比一原版(IIT毕业证)伊利诺伊理工大学毕业证成绩单专业办理
 
Nuclear Power Economics and Structuring 2024
Nuclear Power Economics and Structuring 2024Nuclear Power Economics and Structuring 2024
Nuclear Power Economics and Structuring 2024
 
Immunizing Image Classifiers Against Localized Adversary Attacks
Immunizing Image Classifiers Against Localized Adversary AttacksImmunizing Image Classifiers Against Localized Adversary Attacks
Immunizing Image Classifiers Against Localized Adversary Attacks
 
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
一比一原版(UofT毕业证)多伦多大学毕业证成绩单如何办理
 
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
一比一原版(SFU毕业证)西蒙菲莎大学毕业证成绩单如何办理
 
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdfTop 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
Top 10 Oil and Gas Projects in Saudi Arabia 2024.pdf
 
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...
NO1 Uk best vashikaran specialist in delhi vashikaran baba near me online vas...
 
weather web application report.pdf
weather web application report.pdfweather web application report.pdf
weather web application report.pdf
 
ethical hacking in wireless-hacking1.ppt
ethical hacking in wireless-hacking1.pptethical hacking in wireless-hacking1.ppt
ethical hacking in wireless-hacking1.ppt
 
ethical hacking-mobile hacking methods.ppt
ethical hacking-mobile hacking methods.pptethical hacking-mobile hacking methods.ppt
ethical hacking-mobile hacking methods.ppt
 
J.Yang, ICLR 2024, MLILAB, KAIST AI.pdf
J.Yang,  ICLR 2024, MLILAB, KAIST AI.pdfJ.Yang,  ICLR 2024, MLILAB, KAIST AI.pdf
J.Yang, ICLR 2024, MLILAB, KAIST AI.pdf
 
H.Seo, ICLR 2024, MLILAB, KAIST AI.pdf
H.Seo,  ICLR 2024, MLILAB,  KAIST AI.pdfH.Seo,  ICLR 2024, MLILAB,  KAIST AI.pdf
H.Seo, ICLR 2024, MLILAB, KAIST AI.pdf
 
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdfAKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
AKS UNIVERSITY Satna Final Year Project By OM Hardaha.pdf
 
Gen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdfGen AI Study Jams _ For the GDSC Leads in India.pdf
Gen AI Study Jams _ For the GDSC Leads in India.pdf
 
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
Industrial Training at Shahjalal Fertilizer Company Limited (SFCL)
 
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Hori...
 
Fundamentals of Electric Drives and its applications.pptx
Fundamentals of Electric Drives and its applications.pptxFundamentals of Electric Drives and its applications.pptx
Fundamentals of Electric Drives and its applications.pptx
 
Planning Of Procurement o different goods and services
Planning Of Procurement o different goods and servicesPlanning Of Procurement o different goods and services
Planning Of Procurement o different goods and services
 
HYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generationHYDROPOWER - Hydroelectric power generation
HYDROPOWER - Hydroelectric power generation
 

ALDEP++: An improvement on the ALDEP heuristic via department batching.

  • 1. A N I M P R O V E M E N T O N T H E A L D E P H E U R I S T I C V I A D E P A R T M E N T B A T C H I N G ALDEP++
  • 2. BACKGROUND & MOTIVATION Introduction to ALDEP – • Construction algorithm that randomly chooses the first department • Successively adds more on the basis of closeness ratings of the relationship chart following S pattern Advantages of ALDEP – • simple and not time consuming. • does not need any initial Layout • Since the S-flow method is used in ALDEP, resulting in practical layouts • A greater variety of good options are exposed as compared to other methods
  • 3. BACKGROUND & MOTIVATION Shortcomings of ALDEP - • Shortsightedness - Only considers relationship between immediate previously placed department and each of remaining departments. • Each individual department is Considered for one subsequent department decision • Leads to generation of many suboptimal layouts, not necessarily the most optimal layout. • Blind to constraints (such as disallowing departments with X relation) • Only considers adjacent departments for comparing different layouts via total closeness ratings, there is information lost regarding the total costs. Motivation – • The ALDEP adds one department at a time resulting in blindness to many department relationships • Adding all departments at once makes it a quadratic assignment problem which is computationally difficult as the problem size increases. • ALDEP++ is a method which functions between these two extremes by adding departments in small batches. • ALDEP provides a good starting point for designing a new constructive heuristic algorithm. • New algorithm “ALDEP++” will attempt to counter the shortcomings of ALDEP without abandoning these advantages.
  • 4. EVOLUTION OF ALDEP++ ALDEP++ : The General Approach • Addition of departments in batches • Use of flow * distance as our measurement for what order to place the departments from each batch in the ALDEP++ algorithm. • Slightly modified input compared to ALDEP. • In-between flow matrix is required Benefits of ALDEP++ • Addition of department in batches overcomes the shortsightedness of ALDEP and produce more optimized solutions • Use of flow * distance as a measure , it allows us to obtain total costs of comparable to layouts produced with the MicroCRAFT algorithm while still obtaining high TCR’s (maintaining some of the simplicity of the ALDEP algorithm)
  • 5. MODEL FORMULATION :ALDEP++ Assumptions – • The area of departments is a multiple of sweep width. • The length and width of the building is a multiple of sweep width. • The floor space should be amenable to the serpentine flow. • If there is a tie, departments are chosen/placed randomly. • The shape of department can be changed. • The next department starts where the previous one ends. Batching - • Uses relationship categories with respective scoring weights to determine which departments to add next. • Takes the sum of the relationships to all departments from the previous batch. • Ranks the remaining departments by this relationship sum, the top scoring departments can be determined as the subset to be included in the next batch.
  • 6. MODEL FORMULATION :ALDEP++ Order Determination • Batch of n departments analyze all n! possible orders of placement along the serpentine sweep pattern • Resulting combinations are compared using a flow*distance calculation for the subset of departments consisting of the current batch and the previous batch • The placement order resulting in the lowest total flow*distance is selected and applied • Distances can be measured as either E-DIST or centroid distances, depending on the preferences and limitations of the user Initialization • First n departments are to be placed following the ALDEP algorithm. • The first department is chosen at random, and each successive department is chosen based on the highest relationships with the previously placed department. This process introduces the randomness of the ALDEP process. Iterations • ALDEP++ makes random decisions whenever there are ties. • At the very least, the starting department will be random. • If multiple layouts are tied in the end for highest TCR, total flow * distance is used as the tiebreaker.
  • 7. EXAMPLE MODEL: ALDEP++ Flow-Between Matrix A B C D E F A 0 8 0 4 0 B 0 5 0 2 C 0 1 0 D 6 0 E 4 F Department Size (sq units) A 400 B 800 C 600 D 600 E 800 F 400 Total Facility Size = 60 x 60 units Sweep Width = 20 units Relationship Categories A = 8 E = 6, 5 I = 4 O = 1, 2 U = 0 Scoring Weights A = 64 E = 16 I = 4 O = 1 U = 0 STEP 1 – • Department F randomly selected for the start, and the ALDEP algorithm selected E as the next department as it has the strongest relationship with F • Using a batch size of 2 Initial two departments (first batch) F E
  • 8. EXAMPLE MODEL: ALDEP++ Flow-Between Matrix A B C D E F A 0 8 0 4 0 B 0 5 0 2 C 0 1 0 D 6 0 E 4 F Department Size (sq units) A 400 B 800 C 600 D 600 E 800 F 400 STEP 2 – Second batch: • Calculating the relationship ratings based on the scoring weights, it is determined that A and D are the departments for the second batch. • Because n=2, there are 2!=2 order combinations Department Relationship Rating A 4+0 = 4 B 0+1 = 1 C 1+0 = 1 D 16+0 = 16 F A E D F D E A Alternative 1 Alternative 2
  • 9. EXAMPLE MODEL: ALDEP++ Flow-Between Matrix A B C D E F A 0 8 0 4 0 B 0 5 0 2 C 0 1 0 D 6 0 E 4 F F A E D Alternative 1 Alternative 2 Calculating the sum-product of the E-DIST matrices with the flow matrix gives 490 for order 1 and 364 for order 2. Because we want the minimum flow*distance, we select alternative 2.
  • 10. EXAMPLE MODEL: ALDEP++ Flow-Between Matrix A B C D E F A 0 8 0 4 0 B 0 5 0 2 C 0 1 0 D 6 0 E 4 F F B A E D C Alternative 1 Alternative 2 • This is the last batch, so we can immediately skip to testing the 2 order combinations. • Calculating the sum products gives 603.75 for order 1 and 371.68 for order 2. Therefore we choose order 2, and that is the completed layout. Third Batch: F C A E D B Alternative 1 Alternative 2
  • 11. EXAMPLE MODEL: ALDEP++ • This final layout has a total flow*distance of 1066.7 as compared to the results from the standard ALDEP procedure which results in a higher total of 1132.9. • Under the adjacency scoring method, the ALDEP++ layout receives a total weighted score of 104 as compared to the ALDEP layout which scores only 101. • In this example, ALDEP++ has outperformed ALDEP under both methods of comparison. F B A E D C F C A E D B ALDEP LAYOUT total weighted score – 101 Flow*dist = 1132.9 ALDEP++ LAYOUT Total weighted score – 104 Flow*dist = 1132.9
  • 12. IMPLEMENTATION & RESULTS IMPLEMENTATION – • ALDEP++ was compared with the original ALDEP and MicroCRAFT • All three heuristic algorithms implanted with Matlab code • ALDEP++ was coded using centroid distances primary reason for this decision being making a fair comparison in total cost with the results from MicroCRAFT, which also is based on centroid distance. • Layout problems with appropriate variability generated and the run in these three codes RESULTS : For all cases, our outputs were TCR (total closeness rating) for adjacency based scoring, and centroid distance * flows for the cost based scoring.
  • 13. IMPLEMENTATION & RESULTS Interpretation : • TCR results for the ALDEP++ layouts are generally very close to the TCR results from ALDEP • Comparing case-by-case, it can be seen that ALDEP++ works better when: • The layout is square or close to square. • Department sizes are not highly variable. • Each batch roughly fills one column of the serpentine flow pattern. Total Material Handling Cost Comparison ALDEP++ yields lower costs on average because ALDEP does not consider flow * distance in it’s algorithm, while ALDEP++ does. ALDEP++ gives significantly better TCR than MicroCRAFT. As a result, while ALDEP++ consistently has high TCR, MicroCRAFT results in highly variable TCR’s
  • 14. BENEFITS & CONCLUSIONS Benefits - • ALDEP++ procedure improves on the performance of the layouts generated through adding departments in batches. • In general, we have found that the ALDEP++ algorithm is producing lower cost layouts when compared to the total costs obtained from the ALDEP algorithm. • The layout TCR’s obtained from the ALDEP++ algorithm are not significantly lower than the TCR’s seen from the ALDEP algorithm. • ALDEP++ has the best properties of both MicroCRAFT and ALDEP algorithms. • ALDEP++ also has the advantage of covering one of ALDEP’s flaws, in avoiding undesirable X- relationships. • Batch-approach of ALDEP++ allows it to avoid X-relationships more effectively Conclusions – • ALDEP++ is a feasible heuristic algorithm for generating new layouts • Provides balance between adjacency based TCR scoring and distance * flow total cost scoring • ALDEP++ performs similarly to ALDEP with respect to TCR, and also obtains low-cost layouts similar to the results from MicroCRAFT. • Generally, ALDEP++ works best when the factors allow single batches to fit in one pass of the serpentine flow pattern • As total cost is an important realistic consideration for comparing layouts, ALDEP++ offers facility layouts that are more practically optimal as compared to ALDEP.