Featured Product
This Week in Quality Digest Live
Lean Features
Yoav Kutner
Let salespeople spend more time on customer service, market research, and competitor analysis
Bruce Hamilton
A story of teacher and student
Andrew Peterson
Small manufacturers want robots with more human-like dexterity and self-control
Ryan E. Day
Can lean manufacturing ease the U.S. housing crisis?
Ashley Y. Metcalf
How lean and JIT organizations can withstand supply chain disruptions

More Features

Lean News
Freedom platform connects to any industrial asset to provide automated intelligence related to asset availability, utilization, and continuous improvement
Galileo’s Telescope describes how to measure success at the top of the organization, translate down to every level of supervision
Too often process enhancements occur in silos where there is little positive impact on the big picture
This book is a tool for improvement and benchmarking
Real-time data collection and custom solutions for any size shop, machine type, or brand
Collect measurements, visual defect information, simple Go/No-Go situations from any online device
What continual improvement, change, and innovation are, and how they apply to performance improvement
Incorporates additional functionality and continuing improvements to the product’s existing rich features
Good quality is adding an average of 11 percent to organizations’ revenue growth

More News

James Brewton

Lean

Is It Time (Again) for Office Work Measurement?

An often overlooked component in lean office initiatives

Published: Wednesday, December 7, 2011 - 15:25

Armed with a degree in organizational psychology, I started my career in operational improvement during the early 1970s with a nationally recognized Hartford, Connecticut-based insurance company. As part of the company's productivity services team, my job was to conduct stopwatch time studies in the company's business operations in support of an organizationwide—and very successful—productivity improvement program.

During the next 20 years, the techniques of work measurement in administrative environments (e.g., banks, insurance companies) improved in accuracy and speed of application, although the employee work-measurement performance data were frequently misused by management. The light W. Edwards Deming and his 14 points shed on this issue, and the resulting shift in management focus from productivity to quality and customer satisfaction, caused office work-measurement programs to dwindle. By the end of the 1980s, they were virtually dead.

Having remained in the discipline of administrative operational improvement for my professional career, and for the past several years working as an independent lean office consultant, I have been continually amazed that office work measurement remains a missing key component in today's lean office initiatives. Yes, the need for time studies is often mentioned in articles, books, and presentations on lean offices, but few real examples of office time-study techniques and their benefits are described.

Those of us with a background in office work measurement know there is a great difference among the administrative work measurement techniques available to today's lean office practitioners, especially in how well these techniques create accurate time standards, and analyze and improve work methods (see figure 1).

is it time figure1c
Figure 1:
Administrative work measurement techniques
*Leveled time study

For today's lean office initiatives, there could be no better time to reintroduce predetermined time systems (PDTS) to office work measurement. Although the benefits are many, three clearly stand out:
1. Accurate measurement of administrative process cycle time
2. Accurate determination of administrative process staffing requirements
3. "Micro"-leaning of administrative processes

Measuring administrative cycle time with office PDTS

Figure 2 illustrates an example of an office predetermined time work-measurement system applied to the current state of the administrative activity, enter credit report requests.

is it time figure2b
Figure 2:
Office predetermined time system example

The activity analysis used in measuring administrative process cycle time with an office PDTS is comprised of five components:
• Specific cycles of action (i.e., motions and mental decisions) that make up the administrative process
Predetermined time code that represents each cycle of action performed
Predetermined time value for each cycle of action
• Frequency of occurrence for each cycle of action in an average batch of work
Total time for each cycle of action, calculated by multiplying the lean measurement units (LMU) value for each cycle of action by its frequency of occurrence.

Once the basic data have been analyzed and calculated, cycle time for the work unit being processed (a credit report request in our example) is determined through a set of simple calculations.

First, the total time for all cycles of action is summarized, resulting in total time (the LMU for this office PDTS). Next, total time is divided by the average batch size (50 in our example) for the activity, resulting in the total time per work unit (LMU / request). Finally, total cycle time per work unit is determined by increasing total time per item (request) by 10 percent, and then converting it to hours per item for a later staffing calculation:

(467 LMU + 10% = 514 LMU; 514 LMU / 100,000 LMU per hour = 0.0051 hours per request).

Ten percent is a common time allowance added to PDTS to account for normal unplanned work interruptions like getting supplies, briefly speaking with a supervisor or colleague, trips to the restroom, or getting a cup of coffee. Planned work interruptions (e.g., lunch, breaks, weekly meetings, vacations) are handled in a different way and described later in this article.

Predetermined time systems reflect the time required by the average fully trained worker to complete a cycle of action right the first time.

All of today's recognized administrative predetermined time systems are directly or indirectly based on a common predetermined time system developed in the mid-1940s called "Methods Time Measurement" (MTM). MTM is the most widely used predetermined time system in the world. It was developed by Harold Maynard, Gustave Stegemerten, and John Schwab, industrial engineers at Westinghouse Electric Corp. Its development was spurred by the need for greater consistency in manufacturing work-measurement studies using the stopwatch time-study method.

In its basic form (MTM-1), MTM is comprised of 10 measure categories and more than 800 time values. MTM is defined when a process analyzes a manual operation or method into the basic motions required to perform it, and a predetermined time standard is assigned to each motion. the predetermined standard is determined by the nature of the motion and the conditions under which it is made. Figure 3 illustrates an example from the MTM-1 work measurement system.

is it time figure3
Figure 3: Partial MTM-1 data table for "reach, the basic motion used to move the hand or fingers to a destination or location

MTM time values are expressed as time measurement units (TMU) and represent very small amounts of time. For example, the MTM time to reach for an object like a stapler is 0.00014 hours, 0.0009 minutes, or 0.519 seconds. Working with numbers like these is difficult. To remedy this situation, Maynard, Stegemerten, and Schwab looked for a way to work in whole numbers. The speed of the camera used in filming and measuring the more than 800 different motions in developing the MTM-1 system was 16 frames per second. Each frame had an elapsed time of 0.00001737 hours. Based on this analysis, the MTM developers settled on a 1 TMU value of 0.00001 hours or 100,000 TMU per hour. A TMU conversion table is shown in figure 4 below.

https://lh4.googleusercontent.com/Ehigp5yaPA3OE0Vzf6Nv7ujnUdn2It1R8DTefCbFDT3rGTijjhfvj7cwcAkJoCUFZOz2Wx7xILYQSIyk5eEhg9TzEuS9sHYovOoBQrbYbKZgGPPx1CM
Figure 4: TMU conversion table

The LMU used in the examples in this article is based on and directly equal to the TMU.

Although MTM-1 provides consistently accurate time standards for work activities, it is slow in application. To remedy this, faster versions of MTM-1 have been developed over the years, including special versions called "standard data" for specific work environments (e.g., manufacturing, warehouse and material handling, office) using what is known as the "building block" concept.

Using MTM-1 and the building block approach, an "office" standard data system can be developed that ensures continued standards accuracy but significantly reduces the time for application. An example is shown in figure 5.

is it time figure5
Figure 5:
Building-block approach

In our example in figure 5, using basic MTM requires carefully analyzing 14 different motions to arrive at the correct time for removing an object (e.g., staple remover) from a desk drawer. Office PDTS data reflect the cycles of action normally performed in an office or administrative work environment. With an office PDTS, all that is needed to ensure that an accurate activity time is developed is for the lean office practitioner to recognize the specific cycle of action being performed (e.g., employee obtains a work object from a closed drawer) then locate the corresponding code and time value in the office PDTS data system. With some office PDTS, a few weeks' use can commit the entire work-measurement system to memory by a lean practitioner.

"Leaning" administrative processes with office PDTS

With a detailed analysis of an administrative activity's current state, focus can shift from measurement to elimination of activity waste in the future state. Leaning with administrative PDTS is accomplished by applying one of three "leaning levers" to each cycle of action:
1. Eliminate the cycle of action
2. Reduce frequency of the cycle of action
3. Substitute a less time-consuming cycle of action

Unlike typical lean analyses that focus only on selected nonvalue-added action, office PDTS allow a thorough analysis of every cycle of action (both value added and nonvalue-added) that employees use in completing an activity. Office PDTS also allow the immediate calculation of the impact on time and staffing resulting from the waste reduction recommendations without actually having to implement them (figure 6).

is it time figure6
Figure 6:
Leaning using office PDTS

For example, a likely nonvalue-added "leaning" opportunity in our administrative activity example is reducing the number of steps (Lever 2 in figure 6) that a processor must make in delivering completed batches of work. If the distance could be shortened to 10 steps one way, we can quickly calculate the impact as 600 fewer LMU (or 21.6 fewer seconds) per batch.

With the way office PDTS present the action performed, other "leaning" opportunities quickly present themselves, such as computer applications that automatically place the cursor in the name field ready for text entry (Lever 1) or automatically tab to the next data entry field (Lever 3). Implementing just these waste-reduction levers results in 4,250 fewer LMU (153 fewer seconds per batch), a 21.8-percent time savings. Additional analysis might identify opportunities for waste reduction in value-added cycles of action (where most of the time is consumed) such as having the city, state, and zip code data field automatically filled when the telephone number is entered by the processor.

Determining staffing requirements using office PDTS

One of lean's most useful contributions is the formula for determining the staff required for meeting customer and cycle-time demands for a selected process:

Cycle time / Takt time

Cycle time is defined as the time required to process one work item (e.g., request) in an activity, process or entire value stream. Takt time is defined as the production pace that an activity, process, or value stream must maintain to meet customer workload demand.

One of the most difficult challenges for today's administrative organizations attempting to implement lean is to accurately determine the number of full-time equivalent employees (FTE) required to effectively meet customer demand. Too few FTE results in the inability to meet customer demand. Too many FTE results in excess cost for the organization. Both conditions are classic waste. Without an accurate staffing model, organizations must often go through several stages of staffing adjustments until they get it right, all the while continuing to deliver poor customer service and incur excess expense.

To meet this challenge, office PDTS provide a comprehensive approach for defining and measuring cycle time and takt time to ensure required staffing calculations are accurate from the beginning.

In office work measurement, cycle time is defined as the time it should take an average well-trained employee to produce one work item in an activity, process, or value stream. It includes setup and wrap-up actions but excludes delay or "wait" time during work processing.

As shown in figure 2, cycle time calculation is a two-step process. First, a "base" cycle time (i.e., cycle time without normal work interruptions) for producing one work item is calculated. In our current-state example, 467 LMU / Request has been calculated as the base cycle time for this administrative activity.

Once the base cycle time has been calculated, it is increased by a miscellaneous work interruption (MWI) allowance. In our example (figure 2) total cycle time becomes 514 LMU / Request (or 18.49 seconds) after applying MWI. At this point we have half the data we need to accurately calculate the staffing required for this activity.

The next task is to calculate takt time. Most articles, books, and presentations on lean do a cursory job of explaining how to accurately calculate takt time. Most simply say "determine the basic daily work schedule, subtract lunches and breaks, then divide this result by the average number of customer requests per day to arrive at takt." Unfortunately, this is an oversimplification and inaccurate method for calculating true takt time and, again, can result in an organization having to go through several adjustments to accurately balance process staff to workload.

Takt time is calculated as:

Net available time / Customer workload demand

Calculating workload demand for an administrative activity is usually a relatively simple process for most administrative activities. Nevertheless, it should be calculated as accurately as possible. For most administrative activities where work volumes remain relatively stable, calculating average weekly or monthly work volumes is usually sufficient.

Calculating net available time is a somewhat more complicated but essential calculation for accurate staffing. An effective office PDTS should apply four critical work schedule adjustments in arriving at an accurate net available time:
1. Breaks/lunch/meeting adjustment
2. Attendance adjustment
3. Unmeasured work adjustment
4. Performance adjustment

Breaks/lunch/meeting adjustment

This adjustment accounts for employee unavailability for work due to scheduled lunches, breaks, and recurring weekly meetings. For example, if employees are allowed one hour for lunch and breaks each day of a five-day work week with no recurring meetings, the weekly breaks/lunch/meetings adjustment would be 5.63 hours per week:

45 hours* – 5 hours = 40 hours; 5 hours / 40 hours = 12.5%; 12.5 % × 45 hours = 5.63 hours
(*work schedule in example = 8 a.m.–5 p.m., 5 days per week = 45.0 hours)

Attendance adjustment

This adjustment accounts for unavailable resources due to staff members absent and unavailable to perform work. The attendance adjustment can be calculated by determining the number of personal days off (PDO) for employees performing the target activity, process, or value stream. PDO includes vacation days, sick days, and other earned personal time off.

For example, if the average annual PDO is 15 days for employees performing a specific administrative activity, the attendance adjustment would be 2.88 hours per week (250 workdays per year – 15 PDO days per year = 235 available workdays per year; 15 scheduled workdays / 235 available workdays = 6.4%; 6.4% × 45 hours per workweek = 2.88 hours).

Unmeasured work adjustment

This adjustment accounts for staff performing nonstandard work (e.g., special projects, nonroutine work) during the scheduled work period. Experience in administrative work environments shows that the typical range for unmeasured work is 5 to 10 percent. Using 5 percent would result in an unmeasured work adjustment of 2.39 hours per week (100% – 5% = 95%; 5% / 95% = 5.3%: 5.3% × 45 hours = 2.39 hours).

Performance adjustment

The performance adjustment accounts for work that is not completed due to normal fluctuations in work group performance because of staff turnover. One hundred percent is considered the work pace of an average well-trained employee for MTM-based PDTS. Eighty five to 90 percent is the performance range most administrative work teams should be expected to maintain. Using 90 percent as the performance goal for a work team performing an activity, process, or value stream would result in a performance adjustment of 5.00 hours (100% – 90% = 10%; 10% / 90% = 11.1%; 11.1% × 45 hours = 5.00 hours).

Figure 7 shows the results of the adjustments to work schedule in our example administrative activity.


Figure 7: Work schedule adjustments

 

Based on work schedule adjustments, true net available time for this example would be 29.11 hours per week (45 work schedule hours – 15.89 unavailable hours = 29.11 net available hours).

If it was determined that this work activity received 15,000 requests per week, then the resulting takt time for this administrative activity would be 0.0019 hours (29.11 hours per week / 15,000 requests per week = 0.0019 hours). Using typical calculation instructions found in books and articles on lean, net available time would have been significantly (and erroneously) more (45 hours – 5.63 hours for breaks and lunch = 39.37 net available hours); roughly 10 hours more per week than actually available in our example.

With both cycle time and takt time accurately calculated, the staff required to cost effectively meet customer workload demand can now be determined (0.0051 cycle time / 0.0019 takt time = 2.7 FTE).

With an office PDTS, impact on staffing requirements can be immediately known as potential leaning levers can be applied without having to actually implement the changes.

Conclusion

Office PDTS directly address key lean office goals. Among these are:
• Establishing accurate cycle time measures
• Determining accurate staffing requirements for current or future state administrative activities
• Fully identifying, quantifying, and eliminating waste
• Identifying more efficient office layouts
• Developing standard office procedures

Administrative organizations adopting an office PDTS will have a powerful addition to their lean toolkits and experience a significant boost in their return from their lean office programs.

Discuss

About The Author

James Brewton

James Brewton, mangaging director of Lean Office Consultants, is a U.S.-based independent consultant specializing in office/administrative operations improvement. He has 20 years of experience in assisting financial services, insurance companies, government agencies, and industrial company administrative functions in measuring and improving operational value delivery.