Title: Repetitive Scheduling and Lean Construction
1Repetitive Scheduling and Lean Construction
- by
- I-Tung Yang
- Assistant Professor, Chaoyang University of
Technology - Ph.D., University of Michigan, Ann Arbor
2(No Transcript)
3Presentation Outline
- What is Lean?
- Where can Lean Philosophy apply in Construction?
- Classification of repetitive projects
- Work continuity (unforced idleness)
- Limitations of previous techniques
- Proposed modeling elements and algorithm
- Repetitive Project Planner (RP2)
- Case Studies
- Summary and conclusion
- Current progress and future extension
4What is Lean?
- Production philosophy
- Craftsmanship Mass Production Lean
Production - Make more faster Make what is needed when
it is needed - Ultimate purpose eliminate waste (no
value-adding) - Many concepts are common sense but require
implementation. - Where is the value? How to map the value-stream?
- Individual efficiency vs. system efficiency
5Where can Lean Philosophy apply in Construction
- View of planning units
- Discrete activities vs. Continuous flow
- In-process quality control (Andon)
- Result-focused vs. Process-focused
- Benchmarking
- To manage, you have to measure your
performance. - Employee participation
- Centralization vs. Decentralization
- Process transparency
- Visual impact of scheduling
6Where can Lean Philosophy apply in Construction
(Continued)
- Supply chain management
- Long-term, strategic alliance
- Just-in-time delivery (Kanban)
- Increase production reliability
- What can you do instead of What should you do
- Lookahead scheduling
- Shield downstream activities from uncertainty
- Pull/Push scheduling
- Upstream activities push downstream activities
- Downstream activities pull upstream activities
7Some References
- Koskela, L. (1992). Application of the New
Production Philosophy to Construction. Technical
Report 72. Department of Civil Engineering,
Stanford University. - Shingo, S. (1988). Non-stock production.
Productivity Press, Cambridge, Ma. 454 pp. - Womack, J. P., Jones, D. T. (1996). Lean
Thinking. Simon Schuster, New York. 350 pp. - Journal of Construction Research (2002). Special
Issue Lean Construction. Vol. 3, No. 1. 192pp. - International Group of Lean Construction
(IGLC)http//cic.vtt.fi/lean/
8Classification of Repetitive Projects
9Unforced Idleness
- Cause improper scheduling
- Problems
- unproductive waiting
- comeback delay
- relocation cost
- work discontinuity
- morale impact
10State of the Art (I)
- CPM and PERT
- Too big for what it does
- Cannot eliminate unforced idleness
11State of the Art (II)
- LOB and its variations
- Graphical, production diagrams, manual
- Discrete vs. Continuous
- Typical impractical assumptions
- Activities need to be performed in all units
- Production rates and work quantities are
identical in all units
12State of the Art (III)
- Mathematical Optimization Models
- Linear programming, Dynamic programming, Integer
Programming - Additional limiting assumptions
- activities are in series (one predecessor and one
successor) - activities follow the same direction
- Minimization of cost made the model complex
- Other optimization tools neural networks,
genetic algorithm
13State of the Art (IV)
- Simulation Techniques
- Variability in production rates
- Resource interaction
- By themselves do not solve the problem of
unforced idleness
14Limitations of Previous Scheduling Models
- Unrealistic assumptions
- No computational algorithm to generate schedules
- No distinction between units and locations
- Focus on activities only (not links)
- Cannot answer Through what kinds of links are
activities controlled by their predecessors?
15Realistic Repetitive Scheduling Model
- Same crew may work in multiple activities
- One activity may employ multiple crews
- Varied production rates, work quantities
- Activities have multiple predecessors
- Activities may skip units
- Custom progress direction and work sequence
- Ability to specify space buffer or time buffer
between activities - Activities occupy entire areas (blocks)
- Activities occur at locations (bars)
16Objectives
- Realistic
- Necessary modeling elements (activity and link
types) - Universal
- Schedule both discrete and continuous projects
- Practical
- Easily understood graphical schedules
communication - Computational algorithm to automatically generate
schedules (Repetitive Scheduling Method, RSM)
17Units vs. LocationsDiscrete vs. Continuous
18Modeling Elements (I) Activities
- Activity types
- Line
- Block (e.g., excavation, foundation)
- Bar (e.g., culverts, bridges)
Time
Units
Locations
Time
19Production Line (I)
- Production rate (slope)
- Resource Production Rate (e.g., cubic yards/day)
- Unit Production Rate (e.g., units/day) gt unit
duration - Start and finish units (locations)
- Desirable work interruption between units
20Blocks and Bars
Finish Time
Finish Time
Start Time
Start Time
Start Location
Finish Location
Location
21Classification of Links
- What are the types of activities (Line, Block,
and Bar)? - What does the link represent (Tech., Resource,
Continuity)? - What is the measure of the buffer (time versus
distance)? - Where does the link connect?
- Different unit of the same activity (implied in
the algorithm) - Same unit of different activities
- Different unit of different activities
- Is the project continuous or discrete (units
versus locations)? - Which time value does the link connect
(start/finish)?
22Modeling Elements (II) Links
23Time-buffer versus Distance-buffer
24Global versus Local Time-Controlled
25Continuity Relationship
26Cyclic Links
27Computational Algorithm (Stage 1) Simple View
28Computational Algorithm (Stage 2)Pull
Predecessor to Maintain Continuity
- At most one continuity predecessor
- Continuity predecessor set
- Steps (backward calculation)
- Check for cyclic links if yes, skip the activity
- Pull continuity predecessor (h)
- Pull successors of h (re-do forward calculation)
29Pulling Process
30Repetitive Project Planner (RP2)
- 32-bit Window-based program
- Implement the computational algorithm
- Validation tool
- Can it model realistic situations?
- How long it takes to generate graphical schedules
- Are the results reasonable?
- Tested in real-life projects and for educational
purpose
31RP2 (I) Main Window
32RP2 (II) Unit/Location Window
33RP2 (III) Relationship Window
34RP2 Output Graphical Schedule (.jpg or .ps)
35RP2 Output Tabular Report (.xls)
36Case (I) Pipeline Project
37Case (I) Pipeline Project
2 Painters
4 Painters
38Case (II) Highway Rehabilitation
- 8 km, two-lanes
- East- and West-bound
39Case (II) Highway Rehabilitation
- Major repeating activities
- Patch Work
- Dowel bar Retrofit
- Diamond Grinding
- Asphalt
- Other activities
- Traffic Control
- Pavement Painting
- Shoulder Corrugation
40Case (II) Highway Rehabilitation
41Case (III) Four-Story Apartments
42Case (III) Four-Story Apartments
- Floor plan
- Special work sequence (superstructure)
43Case (III) Four-Story Apartments
44Case (III) Four-Story Apartments
45Summary and Conclusion
- Realistic
- - Necessary modeling elements
- Universal
- - Schedule both discrete and continuous projects
- Practical
- - Easily understood graphical schedules
- - Automated generation of schedules
46Current Progress and Future Extension
- Being able to identify critical activities and
controlling sequence - Field experiments of RP2
- Cost-loading feature
- Input for further optimization (buffer time,
resource combinations, etc.) - Input for stochastic sensitivity analysis
- Customized grid lines and text fonts
47Critical Activities and Controlling Sequence
48Questions and Comments?