W5 – ABM- Assessment of Productivity and Learning Curve for Trestle Barrier Wall Construction at SPJ – 12B

  1. PROBLEM DEFINITION

Construction of Reinforced Concrete Barrier wall has commenced along trestle deck at SPJ – package 12B. Work scope includes erection of steel formwork barrier, installation of cast in items and steel angles, casting of concrete and stripping of forms. Rebar is already installed by separate crews working in advance.

Casting works commenced on the 12th August with an initial target completion date of 1st December  (15 weeks) however Subcontractor has constructed only 30Lm of wall since the 12th August and it appears that the Subcontractor will struggle to achieve his average target production of 90Lm/week however subcontractor is still very much in a learning curve phase and work force levels have not reached planned targets as such, rate of production (units / week) is expected to increase over time as crews become more familiar and numbers increase.

Works scope / details are as follows

We must now determine if the Subcontractor will be able to meet the programme using an analysis of his production to date whilst making  allowances for anticipated increases in productivity obtained through repetition and learning curve phases.

2. FEASIBLE ALTERNATIVES

Using data from the last 2 weeks and with consideration of improvements expected as more units are produced, we will assess the cumulative average time to complete each unit/ wall section and compare against his planned figures.

From there will look at whether the current programme and manning levels are realistic and will result in the barrier being completed in time. From this we develop an action plan based upon the following alternatives;

  1. Do nothing. Barrier can be completed by the specified dates
  2. Increasing manpower in current crews (current planned – 10 men)
  3. Increasing overtime / hours worked each day
  4. Extending the programme/ time for completion.
  5. Adding a separate crew and operating with 2 work fronts

3. DEVELOPMENT OF OUTCOMES

 Do Nothing. Production of crews will achieved necessary production and dates

Using the predicted total man hours per section we can estimate the total manhours to complete all 67 sections. Based on current planning the Subcontractor will take 8,643 hours or 90 working days to complete the works.

Extending Manpower in current crews

Increasing manpower may be considered up to a point, there will come a point that more men will not increase production and may in fact reduce output as resources become more and more inefficient. Subcontractor plans to use a 10 man crew to erect formwork, install cast in items, cast concrete and then strip.

Increasing Overtime / hours worked each day

Similar to manpower increases, crews will lose productivity when hours worked are extended.

Standard working hours on site are already 7am to 6pm Monday to Saturday and 7am to 4pm (58 hours/week).

These working hours are considered standard for Malaysia construction industry and not considered excessive. Labour and fatigue guidelines required under projects head contract stipulates that workers cannot work on average more than 75 hours / week.

Whilst previous estimates of productivity losses due to excessive overtime are based on 40 hours weeks (refer to November, 2001 issue of The Revay Report “Calculating Loss of Productivity Due to Overtime Using Published Charts – Fact or Fiction” by Regula Brunies, FPMI, CCC, CQS and Zey Emir, P.Eng, MBA Revay and Associates Limited ), It is proposed to adjust for local standards but maintain similar productivity loss increments as the quantity of worked hours increase.

  • going from a 58 hour work week to a 70 hour workweek we lose on average 10% productivity;
  • going to a 75 hour workweek we drop 17% to only 83% from the base productivity and
  • Max hours worked will be capped at 75hours/week as per contract.

Extending the programme/ time for completion.

The current target date for completion is the 1st December however prior to this date, 75% of the entire trestle length is to be handed over to Mechanical contractor on the 10th November. i.e 1330Lm x 75% = 998Lm by 10th November.

These works were provided to the Subcontractor as a variation to the main scope and do not include penalties for late completion. Planned target dates are included within the variation order to which the Subcontractor has accepted.

In the event the subcontractor does not complete at least 75% of the planned total by the 10th November, it is likely that the interface issues with the mechanical contractor will arise when conveyor works commence however a detailed assessment of co ordination and interface issues will need to be considered before accepting or discounting this option.

Adding an additional Crew

The inclusion of an additional work crew may be considered where the above 3 options are not determined to be unfeasible however such an option would require the following considerations;

  • Fabrication of new steel formwork system (Minimum 4-6 weeks)
  • Recruitment of new team and skilled foreman (timeline unknown)
  • Learning phase allowances for new crew

4. SELECTION CRITERIA

The option which results in completion of at least 75% of the trestle (or as close to) and there after the earliest completion date for the entire length of the barrier wall. Dates are as follows,

  • 75% completion by the 10th November 2017
  • 100% completion by the 1st December 2017

 5. ANALYSIS OF THE ALTERNATIVES  

The Learning curve analysis will be based on crews achieving an 80% learning curve slope and will assess the following aspects;

  • Manhours it will take to cast the last unit (unit 67)
  • The total manhours required to assemble all 67 units
  • The estimated cumulative average manhours for all 67 units

We will commence Assuming a proportional decrease in barrier construction time for output units between doubled quantities, the total manhours needed to construct the last section of barrier wall is calculated by;

 

 From the above table and graph, the cumulative average man hours for the construction of each wall barrier section is 172.15 hours however the planned weekly average man hours for the barrier is estimated as only 129.9 man hours.

This implies that over the 67 units scheduled for construction, there will be a deficit in man hours of (172.15 – 129.9) x 67 units = 2830.75 man hours, which equates to an additional time of 4.88 weeks ( 30 days) of production based upon an 10 man crew working a total of 580 hours / week.

Another way to assess the delay is to consider the SPI. Refer to following table which considers SPI and determines the total forecast duration to be an additional 31 days.

Based on the above, option 1 of the feasible alternatives i.e Do nothing  cannot be considered and an action plan must be developed to determine the correct solution.

In next blog we will analyse and assess alternatives 2 and 3 to increase production levels and satisfy current TARGET dates.

Final Blog will assess alternative 4 – Extend programme and compare all alternatives with final selection of action

Performance Monitoring

Subcontractor will be monitored for the next 2 weeks to further assess their conformance with assumed  learning curve parameters of 80%.

References

  1. Sullivan, G. W., Wicks, M. E., & Koelling, C. P.(2014). Engineering economy 16th Edition. Chapter 3 – Learning and Improvement., pp.110-112.
  2. Humpreys, G. C. (2011). Project Management Using Earned Value (2nd ed.). Chapter 22 Learning Curves. Humpreys & Associates, Inc.
  3. Chapter 11.5 Managing Project Data Basis  – Guild of project controls compendium and reference (CaR) | Project Controls – planning, scheduling, cost management and forensic analysis (Planning Planet). Retrieved from http://www.planningplanet.com
  4. Chapter 9.5 Project Performance Forecasting – Guild of project controls compendium and reference (CaR) | Project Controls – planning, scheduling, cost management and forensic analysis (Planning Planet).  Retrieved from http://www.planningplanet.com
 

1 thought on “W5 – ABM- Assessment of Productivity and Learning Curve for Trestle Barrier Wall Construction at SPJ – 12B”

  1. EXCELLENT case study, Tony and you did a great job setting it up not only for this week but also for a couple of more blogs. That is working smart!!!

    Not sure if you noticed, but Engineering Economy and Humphrey’s use two slightly differnt approaches to calculating Learning Curve? Maybe at some point you can compare the two methods and see if there is any significant difference in the outcomes?

    There is another method that you could also use and that is the “Line of Balance” method. That would work great if you end up adding another crew OR considering working more than one SHIFT. This example of Line Of Balance by Dr. Ahmed H. Elyamany https://goo.gl/yDu2xB is another “Feasible Alternative” you may want to explore as well.

    Keep up the great work and looking forward to more interesting and challenging problems like this one.

    BR,
    Dr. PDG, Jakarta

     

Leave a Reply

Your email address will not be published. Required fields are marked *