Scribd
Upload
31 views10 pages

GP113 Systems 4

Manufacturing systems lecture note 4

Uploaded by

Ravindu Ashan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
31 views10 pages

GP113 Systems 4

Manufacturing systems lecture note 4

Uploaded by

Ravindu Ashan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

GP113: Fundamentals of Manufacture

Section 08
Manufacturing Planning & Control

Chanaka Senanayake
Department of Manufacturing & Industrial Engineering 1
Recall: Scheduling
Job data
(e.g., 𝑝𝑗 , 𝑑𝑗 )

Single Machine
(e.g. CNC Machining Center)

Scheduling Gantt chart of Parameters Performance


rules schedule of interest measures

σ 𝐶𝑗 σ 𝐿𝑗
FCFS, EDD M1 Job 2 Job 3 Job 1
𝐶𝑗 , 𝐿𝑗 ,
σ𝑗 σ𝑗

How to select the best scheduling rule? 2


Objectives: For selecting an optimal scheduling rule

• Minimize average Completion Time of jobs


σ𝐶𝑗
Min
σ𝑗

e.g. Min (C1+C2+C3+C4+C5)/5

• Minimize average Lateness of jobs


σ𝐿𝑗
Min
σ𝑗

e.g. Min (L1+L2+L3+L4+L5)/5

• Minimize the maximum Lateness/Tardiness


Min 𝐿𝑚𝑎𝑥 or Min 𝑇𝑚𝑎𝑥 where 𝑇 = max(0, 𝐿)
Notation: Job Scheduling Problems

α|β|

Objective
Machine environment

Processing characteristics

Examples

1||σ𝐶𝑗 - Single machine scheduling problem to minimize sum of completion times.

𝐽3 𝑟𝑗 𝐿𝑚𝑎𝑥 - Three machine job-shop with different job release times to minimize
maximum lateness
Gantt Chart: Multiple Resources

J1,1 J2,3 J4,2 J3,3


Resource/Machine

J2,1 J1,2 J4,1 J3,2

J3,1 J2,2 J1,3 J4,3

Fig 1: A Gantt chart showing 4 jobs scheduled on 3 machines


Finding the Optimal Sequence

Theorem: SPT rule is optimal for 𝟏||σ𝑪𝒋

Proof by contradiction: Lets assume a job schedule which is not arranged


according to SPT is optimal. Call it schedule S Should be at least
two adjacent jobs J
and K where 𝒑𝑱 > 𝒑𝑲

J K

K J

Interchange jobs 𝐽 and 𝐾 and create schedule S’


Theorem: SPT rule is optimal for 𝟏||σ𝑪𝒋

J K

K J

Calculate the sum of completion times for both schedules:

σ 𝐶𝑗 (𝑆) = 𝐶1 + 𝐶2 + ⋯ 𝑡 + 𝑡 + 𝑝𝐽 + 𝑡 + 𝑝𝐽 + 𝑝𝐾 + ⋯
σ 𝐶𝑗 𝑆 ′ = 𝐶1 + 𝐶2 + ⋯ 𝑡 + 𝑡 + 𝑝𝐾 + 𝑡 + 𝑝𝐽 + 𝑝𝐾 + ⋯

Since 𝒑𝑱 > 𝒑𝑲 , σ 𝐶𝑗 (𝑆) > σ 𝐶𝑗 𝑆 ′

This contradicts the optimality of S and completes the proof


Some more scheduling rules

Jobs may be allocated weights, 𝑤𝑗 , that represent the cost or value


added.

The longer a high cost job stays in the system, the costlier it is.

In this case, a suitable objective may be to minimize the weighted


completion time

𝟏||σ𝒘𝒋 𝑪𝒋
Example: Weighted completion time

The following data is provided for a single machine scheduling


problem. What is the optimal sequence to minimize the sum of
weighted completion times?

Job 1 2 3 4 5
Processing time, 𝒑𝒋 , in days 5 7 8 4 6
Due date, 𝒅𝒋 , in days 10 20 28 8 15
Cost factor (cost per unit time 12 25 15 10 5
spent in the system)

1|| σ 𝑤𝑗 𝐶𝑗
Homework

Read the note and answer the Example Question

Prove that EDD is optimal for 𝟏||𝑳𝒎𝒂𝒙 using the


adjacent pairwise interchange method

You might also like