22
Presented by: George Brindle, P.Eng, MBA VP Facilities Proving Something Big Partnered with: Janaka Ruwanpura, Ph.D., PQS Canada Research Chair, Professor and Director of Project Management March 8, 2012 Time & Motion: Improving Construction Productivity

Time & Motion - Laricina Energy Ltd. · PDF file• Supporting Time: discussions, toolbox meetings, safety etc. ... supply and demand metrics for oil and natural gas; the impact of

  • Upload
    vobao

  • View
    215

  • Download
    2

Embed Size (px)

Citation preview

Presented by: George Brindle, P.Eng, MBAVP Facilities

Proving Something Big

Partnered with: Janaka Ruwanpura, Ph.D., PQSCanada Research Chair, Professor

and Director of ProjectManagement

March 8, 2012

Time & Motion:Improving Construction Productivity

March-08-12 Proving Something Big

2

Outline

• What’s in it for suppliers and other partners?

• We all get better at what you do!

Productivity

Tool Time

Measurement & Report

Opportunities

• Why focus on ‘productivity’?

• Measuring what matters

• Understanding Tool Time in Oil Sands

Time & Motion • What is a Time & Motion study?

• UofC Research Team

• How it works?

• Interfaces and workflow

• Protecting the privacy of workersPrivacy Protection

Objectives • Objectives of T&M’s

March-08-12 Proving Something Big

3

Time & Motion Study

• Definition:– A Time & Motion (T&M) study is a business

efficiency technique that observes the time andmethods(motions) to perform any type ofwork1.

• Modern model for T&M: Remotelycontrolled video cameras, accessedexclusively by third party researchers.

• Laricina has agreed to participate in aT&M study with the University ofCalgary’s Centre for Project ManagementExcellence:

– Canada Research Chair Dr. JanakaRuwanpura and research assistants

– Construction Visualization and MonitoringCentre

1.- Wikipedia: Archives from Frederick W. Taylor and Frank and Lilian Gilbreth.

March-08-12 Proving Something Big

4

Why Focus on Productivity?

• Productivity is a ratio of production output to what isrequired to produce it (inputs).– What is our output? -> A more efficient project– What is your output? -> More efficient Lm, Tons, etc.

• However, “Productivity” is a complex issue inconstruction because of the interaction of labour,capital, materials and equipment in the output.

• We should all (owners, engineers, contractors andvendors) look for ways to improve constructionproductivity. It is better for business.

5

Why Focus on Productivity?

• Challenges remain in identifying the specific issuesaffecting productivity…

Technical Issues Management Issues

Market ConditionsExternal Issues/factors

Human/Labour Issues

• What if we measured productivity in observation ofthese issues?

• We could manage these and realize savings per Lm, cm,Ton, etc….

March-08-12 Proving Something Big

6

Objectives of Time & Motion

• Monitor construction activities and site operations tocollect tool time for selected work areas of piling, cutting andcapping, setting of modules/tanks/equipment, structuralsupports and interconnects. Interfaces with other activitiescould be considered

• Identify inefficiencies and opportunities forimprovements in the work

• Recommend best practices to improve productivity basedon the observations and consultations with the project team,

• Continue observation and analysis to quantify the impactthat the recommended best practices have on tool time andproductivity.

March-08-12 Proving Something Big

7

Tool Time – Measuring what matters

What is Tool Time?• The amount of time that workers spend in

producing tangible outputs

• Tool time contributes directly to productivity

Non-Tool Time• Supporting Time: discussions, toolbox meetings,

safety etc.

• Ineffective Time: idle time, extra-socializing,searching for tools and materials

March-08-12 Proving Something Big

8

Tool Time

Commercial Construction

Work time50.7%

Looking for materials14.6%

Looking for tools1.5%

Socializing8.6%

Moving7.8%

Instructions2.8%

Idle9.3%

Other4.7%

work time distribution-overall

1.- Hewage, K.N., & Ruwanpura, J.Y. (2006). Carpentry Workers’ Issues and Efficiencies related to ConstructionProductivity in Alberta’s Commercial Construction Projects. Canadian Journal of Civil Engineering, 33(8), 1075-1089.

• 2004 study: Observations of 101 workers from 4construction sites. (Focus: Formwork)1.

March-08-12 Proving Something Big

9

Tool Time

• 2008 studies by University of Calgary show similar tool time1

BEFORE Field Test

Working/Tool Time48.51%

Instruction15.54%

Walking1.79%

Idling15.82%

Socializing/Phoning1.57%

Out of Sight16.77%

In-Office0.00%

Commercial Construction

1.- Construction Monitoring and Visualization Centre reports: aggregated sources.

Tool Time in Oil Sands

March-08-12 Proving Something Big

10

1.- Construction Owners Association of Alberta (COAA) – Source unknown

What about in Oil Sands?

• There are only indicationsbased on commonknowledge

• No recent studies inindustry

• Tool time has beenestimated at +35%1

March-08-12 Proving Something Big

11

Tool Time in Oil Sands

What about in Oil Sands? SAVE$

• This will be the most recentstudy

• This is the first time that thiscamera observation-feedbackmodel is applied to the industry.

• Cost savings can be substantial

• Industry partners can perceivethese gains and becomeindustry leaders: contractors,vendors, suppliers and workersget much better at what they do

March-08-12 Proving Something Big

12

Measurement and Report

Continuous Measurements

Time Activity

08.00 - 08.45 5min Rating

08.45 - 09.00 5min Continuous

09.00 - 09.30 30min Continuous

09.30 – 11.00 5min Continuous

every 30min

11.00 – 11.30 30min Continuous

11.30 - 01.30 5min Continuous

every 30min

01.30 – 02.00 30min Continuous

02.00 – 03.30 5min Continuous

every 30min

03.30 – 04.00 30min Continuous

04.00 – 04.30 5min Rating

• Continuous Observations for along period (minimum 30minutes)

• Continuous Observations for ashort period (5 minutes)

• Five Minute Rating (whether theworker is working or not)

• Observations include descriptionof possible root causes

March-08-12 Proving Something Big

13

Measurement and Report

Monday, October 14, 2011

D i r ect Wor k

T r anspor t i ng

Wai ti ng &I dle

Lat e St ar t s / Ear l y Qui ts

Per sonal Br eaks

T r avel l i ng

T ool / M ater i al sI nst r uct i ons

N

UofC Sample Screenshot (proprietary)

March-08-12 Proving Something Big

14

Measurement and Report

Weekly Summaries - Analysis

March-08-12 Proving Something Big

15

Measurement and Report

• Productivity Improvement Techniques (PIT) Crew– Laricina Team:

• Productivity Improvement Coordinator,

• Technical and Research Coordinators

• assists UofC in communicating findings andrecommendations

– Assist UofC, has no access to video/data

– PIT Crew facilitates implementation of recommendations

Laricina-UofC Interface

• Laricina Site Management shares results with contractorsand workers (obtains feedback)

• Laricina Area Managers approve/reject recommendationsand implement changes

March-08-12 Proving Something Big

16

Measurement and Report

Laricina-UofC Workflow2.- OBSERVATIONS

ANDRECOMMENDATIONS

REPORT

B3.- REPLYRECOMMENDATIO

N FORM

START

1.- UOFCMEASURES

3.- PIT CREWEVALUATES

5.- IMPLEMENTIMPROVEMENT

B2.- DEPARTMENTMANAGERREVIEWS

Path ‘A’ - MINORCHANGE: Call site

manager

Path ‘B’- MAJORCHANGE: Root causeanalysis and solution

B1.- PIT CREWSELECTS RELEVANT

DEPARTMENT

4.- FILLRECOMMENDATI

ON FORM

direct lineA1.- PIT CREW

CALLS SITEMANAGER

direct line

March-08-12 Proving Something Big

17

Privacy Protection

• Strict UofC ethics protocolprotects workers’ privacy– All workers are informed in

orientation on site

– Comments and concerns aredirected anonymously to UofC

– Signs and brochures

• Laricina and contractors complywith Alberta Privacy Act

• Video footage and any personalidentifying information areprotected: UofC releases onlyaggregated data

March-08-12 Proving Something Big

18

Opportunities (something big)

• Example 1: Applying just one best practice…

ProductivityIncrease of 17%

March-08-12 Proving Something Big

19

Opportunities (something big)

• Example 2: Applying a set of best practices

Productivity Increaseof 20%

Can our industry achieve this?What could the 12.5% ToolTime increase and 20%productivity mean to you?

March-08-12 Proving Something Big

20

Opportunities (something big)

• Learning and Teamwork: Participants (Laricina,contractors and workers) can learn more about the projectexecution and cooperate and take action for a commongood– Results are openly shared and discussed with partners

– Feedback is encouraged among workers

– Improvements become a team effort (teams could get creative andcompetitive)

• Improving methods and practices: Contractors improveand are recognized for achievements (as industry leaders)

• Cost-Schedule-Quality : T&M partners realize immediateimprovements (up to 20% productivity improvements)translating in cost, schedule and quality wins.

• Competitive advantage: become ‘best-in-class’!

March-08-12 Proving Something Big

21

This Laricina Energy Ltd. (the “Company”) presentation contains certain forward-looking statements. Forward-looking statements may include, but arenot limited to, statements concerning estimates of exploitable original-bitumen-in-place, predicted recovery factors, steam-to-oil ratios and wellproduction rates, estimated recoverable resources as defined below, expected regulatory filing, review and approval dates, construction and start-uptimelines and schedules, company project potential production volumes as well as comparisons to other projects, statements relating to the continuedoverall advancement of the Company’s projects, comparisons of recoverable resources to other oil sands projects, estimated relative supply costs,potential cost reductions, recovery and production increases resulting from the application of new technology and recovery schemes, estimates ofcarbon sequestration capacity, costs for carbon capture and sequestration and possible implementation schedule for carbon capture and sequestrationprocesses or related emissions mitigation or reduction scheme and other statements which are not historical facts. You are cautioned not to place unduereliance on any forward-looking statements as there can be no assurance that the plans, intentions or expectations upon which they are based will occur.By their nature forward-looking statements involve numerous assumptions, known and unknown risks and uncertainties, both generally and specific, thatcontribute to the possibility that the predictions, forecasts, projections and other forward-looking statements will not occur. Although the Companybelieves that the expectations represented by such forward-looking statements are reasonable, there can be no assurance that such expectations willprove to be correct and, accordingly that actual results will be consistent with the forward-looking statements. Some of the risks and other factors thatcould cause results to differ materially from those expressed in the forward-looking statements contained in this presentation include, but are not limitedto geological conditions relating to the Company’s properties, the impact of regulatory changes especially as such relate to royalties, taxation andenvironmental changes, the impact of technology on operations and processes and the performance of new technology expected to be applied or utilizedby the Company; labour shortages; supply and demand metrics for oil and natural gas; the impact of pipeline capacity, upgrading capacity and refinerydemand; general economic business and market conditions and such other risks and uncertainties described from time to time in the reports and filingsmade with security regulatory authorities, contained in other disclosure documents or otherwise provided by the Company. Furthermore the forward-looking statements contained in this presentation are made as of the date hereof. Unless required by law the Company does not undertake anyobligation to update publicly or to revise any of the included forward-looking statements, whether as a result of new information, future events orotherwise. The forward-looking statements contained in this presentation are expressly qualified by this advisory and disclaimer.

In this presentation “recoverable resources” includes the unrisked arithmetic sum of best estimate contingent resources and prospective resources andproved plus probable reserves as defined in the report of GLJ Petroleum Consultants Ltd. (“GLJ”) regarding certain of Laricina’s properties effectiveDecember 31st, 2010, referred to herein (the “GLJ Report”). “Exploitable OBIP” refers to original-bitumen-in-place that is targeted for development usingthermal recovery technologies. The best and high estimate includes contingent and prospective resources. Contingent resource values have not beenrisked for chance of development while prospective resource values have been risked for chance of discovery but not for chance of development. Thereis no certainty that it will be commercially viable to produce any portion of the contingent resources. There is no certainty that any portion of theprospective resources will be discovered or, if discovered, if it will be commercially viable to produce any portion of the prospective resources. “2P”means proved plus probable reserves and “3P” means proved plus probable plus possible reserves. The SC-SAGD best estimate technology sensitivity(Laricina technology sensitivity) net economic forecasts were prepared on Saleski-Grosmont and Germain-Grand Rapids based on SC-SAGDtechnology and remaining properties based on SAGD/CSS technology“. SC-SAGD” means solvent-cyclic steam-assisted gravity drainage. “CSS” meanscyclic steam stimulation.

Forward-looking statements advisory

March-08-12 Proving Something Big

22

Contact us

Laricina Energy Ltd.

800, 425 – 1st Street SW

Calgary, Alberta T2P 3L8

403-750-0810

www.laricinaenergy.com

[email protected]