Analysis and Design of Asynchronous Transfer Lines as a series of G/G/m queues.

Slides:



Advertisements
Similar presentations
EMBA-2, BUP Major Asad EO Chapter 5: Process Analysis.
Advertisements

Instructor: Spyros Reveliotis homepage: IE4803-C: Advanced Manufacturing Systems Modeling and.
Treatments of Risks and Uncertainty in Projects The availability of partial or imperfect information about a problem leads to two new category of decision-making.
OPSM 301: Operations Management Session 12: Service processes and flow variability Koç University Graduate School of Business MBA Program Zeynep Aksin.
Flowlines: The prevailing layout for High Volume Manufacturing.
1 Variability Basics God does not play dice with the universe. ---Albert Einstein Stop telling God what to do. ---Niels Bohr.
Modeling and Analysis of High Volume Manufacturing Systems.
Introduction to Transfer Lines Active Learning – Module 1
Variability Basics God does not play dice with the universe.
Operations Management Waiting Lines. 2 Ardavan Asef-Vaziri Dec-2010Operations Management: Waiting Lines1  Questions: Can we process the orders? How many.
#11 QUEUEING THEORY Systems Fall 2000 Instructor: Peter M. Hahn
1 Ardavan Asef-Vaziri Sep-09Operations Management: Waiting Lines2  Made-to-stock (MTS) operations  Product is manufactured and stocked in advance of.
Operations Management Waiting Lines. 2 Ardavan Asef-Vaziri Dec-2010Operations Management: Waiting Lines1  Questions: Can we process the orders? How many.
MODELING AND ANALYSIS OF MANUFACTURING SYSTEMS Session 6 SCHEDULING E
1 Performance Evaluation of Computer Networks Objectives  Introduction to Queuing Theory  Little’s Theorem  Standard Notation of Queuing Systems  Poisson.
Lecture 11 Queueing Models. 2 Queueing System  Queueing System:  A system in which items (or customers) arrive at a station, wait in a line (or queue),
7/3/2015© 2007 Raymond P. Jefferis III1 Queuing Systems.
Little´s law Skorkovský ,KPH,ESF.MU
Analysis and Design of Asynchronous Transfer Lines as a series of G/G/m queues: Overview and Examples.
1 Ardavan Asef-Vaziri Sep-09Operations Management: Waiting Lines3  Terminology: The characteristics of a queuing system is captured by five parameters:
WINTER 2012IE 368. FACILITY DESIGN AND OPERATIONS MANAGEMENT 1 IE 368: FACILITY DESIGN AND OPERATIONS MANAGEMENT Lecture Notes #3 Production System Design.
Lesson 11: Solved M/G/1 Exercises
Modeling and Analysis of CONWIP-based Flowlines as Closed Queueing Networks.
Dr. Cesar Malave Texas A & M University
The Simulation Project. Simulation Project Steps a.- Problem Definition b.- Statement of Objectives c.- Model Formulation and Planning d.- Model Development.
1 Chapter 5 Flow Lines Types Issues in Design and Operation Models of Asynchronous Lines –Infinite or Finite Buffers Models of Synchronous (Indexing) Lines.
Management of Waiting Lines McGraw-Hill/Irwin Copyright © 2012 by The McGraw-Hill Companies, Inc. All rights reserved.
1 1 © 2003 Thomson  /South-Western Slide Slides Prepared by JOHN S. LOUCKS St. Edward’s University.
Introduction to Operations Research
Queueing Analysis of Production Systems (Factory Physics)
18 Management of Waiting Lines.
1 Chapters 8 Overview of Queuing Analysis. Chapter 8 Overview of Queuing Analysis 2 Projected vs. Actual Response Time.
yahoo.com SUT-System Level Performance Models yahoo.com SUT-System Level Performance Models8-1 chapter11 Single Queue Systems.
Analysis and Design of Asynchronous Transfer Lines as a series of G/G/m queues.
© Wallace J. Hopp, Mark L. Spearman, 1996, Variability Basics.
OPSM 301: Operations Management Session 19: Flow variability Koç University Zeynep Aksin
Introduction to Uncertainty Simulation of Operations.
Penny Fab WIP=1.
Queuing Models in Operations
Facilities design. Main Topics Discrete vs. Continuous Flow and Repetitive Manufacturing Process vs. Product-focused designs and the other currently used.
1 1 Slide © 2011 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
Modeling and Analysis of CONWIP-based Flowlines as Closed Queueing Networks.
1 1 Slide © 2009 South-Western, a part of Cengage Learning Slides by John Loucks St. Edward’s University.
OPERATING SYSTEMS CS 3530 Summer 2014 Systems and Models Chapter 03.
The Effects of Process Variability 35E00100 Service Operations and Strategy #3 Fall 2015.
1 1 Slide Chapter 12 Waiting Line Models n The Structure of a Waiting Line System n Queuing Systems n Queuing System Input Characteristics n Queuing System.
OPSM 301: Operations Management Session 13-14: Queue management Koç University Graduate School of Business MBA Program Zeynep Aksin
© 2015 McGraw-Hill Education. All rights reserved. Chapter 17 Queueing Theory.
Management of Waiting Lines Copyright © 2015 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent.
Queueing Theory. The study of queues – why they form, how they can be evaluated, and how they can be optimized. Building blocks – arrival process and.
Facilities design. Main Topics Process vs. Product-focused designs and the other currently used variations Technology selection and capacity planning.
McGraw-Hill/Irwin Copyright © 2009 by The McGraw-Hill Companies, Inc. All Rights Reserved. Chapter 18 Management of Waiting Lines.
Effects of Variability. Successive Units on Time Line t How variable is this system ? p. 5.
Modeling and Analysis of High Volume Manufacturing Systems
Lecture 14 – Queuing Networks
IE4803-REV: Advanced Manufacturing Systems Modeling and Analysis Fall 2017 Instructor: Spyros Reveliotis homepage:
An introduction to Factory Physics
Flowlines: The prevailing layout for High Volume Manufacturing
Analysis and Design of Asynchronous Transfer Lines as a series of G/G/m queues: Overview and Examples.
Modeling and Analysis of High Volume Manufacturing Systems
IV-2 Manufacturing Systems modeling
System Performance: Queuing
Variability 8/24/04 Paul A. Jensen
Topics Modeling CONWIP Flowlines as Closed Queueing Networks
Using the queuing-theoretic approximations for the performance of “push” and “pull” production lines to address Design Problems.
Process Analysis “If you cannot describe what you are doing as a process, you do not know what you are doing.” W.E. Deming.
CSE 550 Computer Network Design
Example: Designing an asynchronous prod. line
Presentation transcript:

Analysis and Design of Asynchronous Transfer Lines as a series of G/G/m queues

Topics The negative impact of variability in the operation of Asynchronous Transfer Lines Modeling the Asynchronous Transfer Line as a series of G/G/m queues Modeling the impact of various operational detractors Employing the derived models in line diagnosis Employing the derived models in line design

Asynchronous Transfer Lines (ATL) W2W3 TH B2B3 W1 TH B1M1 M2M3 Some important issues: What is the maximum throughput that is sustainable through this line? What is the expected cycle time through the line? What is the expected WIP at the different stations of the line? What is the expected utilization of the different machines? How does the adopted batch size affect the performance of the line? How do different detractors, like machine breakdowns, setups, and maintenance, affect the performance of the line?

Analyzing a single workstation with deterministic inter-arrival and processing times TH B1M1 Case I: t a = t p = 1.0 t WIP ArrivalDeparture TH = 1 part / time unit Expected CT = t p

Analyzing a single workstation with deterministic inter-arrival and processing times TH B1M1 Case II: t p = 1.0; t a = 1.5 > t p t WIP ArrivalDeparture TH = 2/3 part / time unit Expected CT = t p Starvation!

Analyzing a single workstation with deterministic inter-arrival and processing times TH B1M1 Case III: t p = 1.0; t a = 0.5 WIP TH = 1 part / time unit Expected CT   t ArrivalDeparture 2 3 Congestion!

A single workstation with variable inter-arrival times TH B1M1 Case I: t p =1; t a  N(1,0.1 2 ) (c a =  a / t a = 0.1) t ArrivalDeparture 2 3 WIP TH < 1 part / time unit Expected CT  

A single workstation with variable inter-arrival times TH B1M1 Case II: t p =1; t a  N(1,1.0 2 ) (c a =  a / t a = 1.0) TH < 1 part / time unit Expected CT   t ArrivalDeparture 2 3 WIP

A single workstation with variable processing times TH B1M1 Case I: t a =1; t p  N(1,1.0 2 ) ArrivalDeparture TH < 1 part / time unit Expected CT   t WIP

Remarks Synchronization of job arrivals and completions maximizes throughput and minimizes experienced cycle times. Variability in job inter-arrival or processing times causes starvation and congestion, which respectively reduce the station throughput and increase the job cycle times. In general, the higher the variability in the inter-arrival and/or processing times, the more intense its disruptive effects on the performance of the station. The coefficient of variation (CV) defines a natural measure of the variability in a certain random variable.

The propagation of variability B1M1 TH B2M2 Case I: t p =1; t a  N(1,1.0 2 )Case II: t a =1; t p  N(1,1.0 2 ) t WIP t W1W2 W1 arrivalsW1 departuresW2 arrivals

Remarks The variability experienced at a certain station propagates to the downstream part of the line due to the fact that the arrivals at a downstream station are determined by the departures of its neighboring upstream station. The intensity of the propagated variability is modulated by the utilization of the station under consideration. In general, a highly utilized station propagates the variability experienced in the job processing times, but attenuates the variability experienced in the job inter- arrival times. A station with very low utilization has the opposite effects.

The G/G/1 model: A single-station Modeling Assumptions: Part release rate = Target throughput rate = TH Infinite Buffering Capacity one server Server mean processing time = t e St. deviation of processing time =  e Coefficient of variation (CV) of processing time: c e =  e / t e Coefficient of variation of inter-arrival times = c a TH B1M1

An Important Stability Condition Average workload brought to station per unit time: TH·t e It must hold: Otherwise, an infinite amount of WIP will pile up in front of the station. TH B1M1

Performance measures for a stable G/G/1 station Server utilization: Expected cycle time in the buffer: (Kingman’s approx.) Expected cycle time in the station: Average WIP in the buffer: (by Little’s law) Average WIP in the station: Squared CV of the inter-departure times: TH B1M1

Remarks For a station with variable job inter-arrival and/or processing times, utilization must be strictly less than one in order to attain stable operation. Furthermore, expected cycle times and WIP grow to very large values as u  1.0. Expected cycle times and WIP can also grow large due to high values of c a and/or c e ; i.e., extensive variability in the job inter- arrival and/or processing times has a negative impact on the performance of the line. In case that the job inter-arrival times are exponentially distributed, c a =1.0, and the resulting expression for CT q is exact (a result known as the Pollaczek-Kintchine formula). The expression for c d 2 characterizes the propagation of the station variability to the downstream part of the line, and it quantifies the dependence of this propagation upon the station utilization.

Performance measures for a stable G/G/m station Server utilization: Expected cycle time in the buffer: Expected cycle time in the station: Average WIP in the buffer: Average WIP in the station: Squared CV of the inter-departure times: M1 B TH M2 Mm

Analyzing a multi-station ATL TH Key observations: A target production rate TH is achievable only if each station satisfies the stability requirement u < 1.0. For a stable system, the average production rate of every station will be equal to TH. For every pair of stations, the inter-departure times of the first constitute the inter- arrival times of the second. Then, the entire line can be evaluated on a station by station basis, working from the first station to the last, and using the equations for the basic G/G/m model.

Operational detractors: A primal source for the line variability Effective processing time = time that the part occupies the server Effective processing time = Actual processing time + any additional non-processing time Actual processing time typically presents fairly low variability ( SCV < 1.0). Non-processing time is due to detractors like machine breakdowns, setups, operator unavailability, lack of consumables, etc. Detractors are distinguished to preemptive and non- preemptive. Each of these categories requires a different analytical treatment.

Preemptive operational detractors Outages that take place while the part is being processed. Some typical examples: –machine breakdowns –lack of consumables –operator unavailability

Modeling the impact of preemptive detractors X = random variable modeling the natural processing time, following a general distribution. t o = E[X];  o 2 =Var[X]; c o =  o / t o. T = random variable modeling the effective processing time = where U i = random variable modeling the duration of the i-th outage, following a general distribution, and N = random variable modeling the number of outages during a the processing of a single part. m r =E[U i ];  r 2 =Var[U i ]; c r =  r / m r Time between outages is exponentially distributed with mean m f. Availability A = m f / (m f +m r ) = percentage of time the system is up. Then, t e = E[T] = t o / A or equivalently r e = 1/t e = A (1/t o ) = A  r o

Breakdown Example Data: Injection molding machine has: 15 second stroke (t o = 15 sec) 1 second standard deviation (s o = 1 sec) 8 hour mean time to failure (m f = sec) 1 hour repair time (m r = 3600 sec) Natural variability c o = 1/15 = (which is very low)

Example Continued Effective variability: Which is very high!

Example Continued Suppose through a preventive maintenance program, we can reduce m f to 8 min and m r to 1 min Which is low! (the same as before)

Non-preemptive operational detractors Activities that may take place between the processing of two consecutive parts. Some typical examples: –setups –preventive maintenance –operator breaks

Modeling the impact of non-preemptive detractors X = random variable modeling the natural processing time, following a general distribution. t o = E[X];  o 2 =Var[X]; c o =  o / t o. N S = average number of parts processed between two consecutive setups It is also assumed that the number of parts between two consecutive setups follows a geometric distribution, which when combined with the previous bullet, it implies that probability for a setup after any given job = 1/ N S. Z = random variable modeling the duration of a setup t S = E[Z];  S 2 = Var[Z] S = random variable modeling the setup time experienced by any given job = T = random variable modeling the effective processing time = X+S Then, E[S] = t S / N S ; Var[S] = (  S 2 / N S ) + t S 2 ((N S -1) / N S 2 ); t e = E[T] = t o +t S / N S ; ;

Setup Example Data: –Fast, inflexible machine: (2 hr setup every 10 jobs) –Slower, flexible machine: (no setups) No difference!

Setup Example (cont.) Compare mean and variance –Fast, inflexible machine – 2 hr setup every 10 jobs –Slower, flexible machine – no setups Conclusion: Flexibility can reduce variability.

Setup Example (cont.) New Machine: Consider a third machine same as previous machine with setups, but with shorter, more frequent setups Analysis: Conclusion: Shorter, more frequent setups induce less variability.

Example:employing the developed theory for diagnostic purposes M1BM2 t o1 =19 min c o1 2 =0.25 m f1 =48 hrs m r1 =8 hrs MTTR ~ expon. t o2 =22 min c o2 2 =1.0 m f2 =3.3 hrs m r2 =10 min MTTR ~ expon. C a 2 = parts Desired throughput is TH = 2.4 jobs / hr but practical experience has shown that it is not attainable by this line. We need to understand why this is not possible.

Diagnostics example continued: Capacity analysis based on mean values

Diagnostics example continued: An analysis based on the G/G/m model i.e., the long outages of M1, combined with the inadequate capacity of the interconnecting buffer, starve the bottleneck!

Example: ATL Design Need to design a new 4-station assembly line for circuit board assembly. The technology options for the four stations are tabulated below (each option defines the processing rate in pieces per hour, the CV of the effective processing time, and the cost per equipment unit in thousands of dollars).

Example: ATL Design (cont.) Each station can employ only one technology option. The maximum production rate to be supported by the line is 1000 panels / day. The desired average cycle time through the line is one day. One day is equivalent to an 8-hour shift. Workpieces will go through the line in totes of 50 panels each, which will be released into the line at a constant rate determined by the target production rate.

A baseline design:Meeting the desired prod. rate with a low cost

Reducing the line cycle time by adding capacity to Station 2

Adding capacity at Station 1, the new bottleneck

An alternative option:Employ less variable machines at Station 1 This option is dominated by the previous one since it presents a higher CT and also a higher deployment cost. However, final selection(s) must be assessed and validated through simulation.