Performance audit, Department of Transportation, construction management

              PERFORMANCE AUDIT
DEPARTMENT OF TRANSPORTATION
CONSTRUCTION MANAGEMENT
Report to the Arizona Legislature
By the Auditor General
October 1987
87- 11
DOUGLAS R NORTON. CPA
AUDITOR GENERAL
STATE OF ARIZONA
OFFICE OF THE
AUDITOR GENERAL
LINDA J. BLESSING, CPA
DEPUTY AUDITOR GENERAL
October 30, 1987
Members of the Arizona Legislature
The Honorable Evan Mecham, Governor
Mr. Charles L. Miller, Director
Arizona Department of Transportation
Transmitted herewith is a report of the Auditor General, A Performance Audit
of the Arizona Department of Transportation, construction management
function. This report is in response to a July 26, 1985, resolution of the Joint
Legislative Oversight Committee.
The report addresses two areas in which ADOT can improve its construction
staff management. We found that ADOT has not adequately controlled
construction staffing levels, resulting in overstaffing that has cost the state an
estimated $ 7.4 million over the last four years. In addition, ADOT's
Construction Engineering Manpower Management System-- which was designed
to help control construction staffing levels-- is not being utilized effectively
and needs greater management support.
My staff and I will be pleased to discuss or clarify items in the report.
Respectfully submitted,
J e ~ x & Do as R. Norton
~ u d i t o rG eneral
Staff: William Thomson
Peter Francis
Arthur Heikkila
Gregg Halemba
Joseph Moore
Lucinda Trimble
2700 NORTH CENTRAL AVE 0 SUITE 700 0 PHOENIX, ARIZONA 85004 0 ( 602) 255- 4385
SUMMARY
The Office of the Auditor General has conducted a performance audit of the Arizona
Department of Transportation ( ADOT), construction management function in
response to a July 26, 1985, resolution of the Joint Legislative Oversight Committee.
This performance audit is one in a series of audits on ADOT and was conducted as
part of the Sunset Review set forth in Arizona Revised Statutes $ 941- 2351 through
41 - 2379.
AD 0 T oversees highway construction of all interstate, primary and secondary
roadways to ensure that specifications are met and that contractors f u l f i l l their legal
obligations. To perform the oversight function, ADOT has approximately 600
construction staff located throughout the State. In recent years, ADOT has also
contracted with private sector engineering consulting firms to provide oversight on
some projects.
ADOT Has Not Adequately Controlled Construction Staffing Levels,
Resulting In Overstaffing That Has Cost The State
An Estimated $ 7.4 Million Over The Last Four Fiscal Years ( see pages 9 through 26)
Over the last four years ADOT has overstaffed its construction function by 21 to 101
unnecessary staff annually, costing the State approximately $ 7.4 million. Our
estimate of overstaffing is based on analysis of data on three variables: 1)
construction engineering ( CE) costs, 2) the substantial variations in staff
productivity, and 3) excessive amounts of non construction time charged by staff.
Our estimates were further supported by Auditor General staff field visits to
construction orgs and ADOT's own field reviews which provided several examples of
overstaffing.
0 A District Two official acknowledges his district will be overstaffed by an
average of 44 employees over the next 12 months.
0 District Four maintains a permanent construction staff ( averaging approximately
70 staff) year- round, although construction activities vary during the year due to
climatic conditions. As a result, one resident engineer stated his org was
overstaffed from eight to 14 positions for nine months of the year. Further,
ADOT field reviews recently reported a surplus of 22 out of an average of 47
staff in two District Four orgs.
ADOT field reviews also found significant amounts of excess hours charged to
the survey function for several projects. ADOT found one project's survey
charges overrunning estimates by 240 to 315 percent. Another project overran
estimates by 6,000 hours. Central office staff attributed the overruns to staff
charging excess time to this function.
ADOT's construction program will expand in three of the next five years. Before
increasing staff size, ADOT should first consider making more use of consultant
engineers, contracting out more of the construction engineering functions ( such as
surveying) and making better use of its own staff resources. These actions will help
prevent staffing for peak workloads and the resultant overstaffing once the peaks
have passed.
Because ADOT has failed to properly adjust past staffing levels to meet fluctuations
in workload, we recommend that the Legislature consider setting and monitoring 4
construction staffing levels, as is done in several other states. We also recommend a -
follow- up performance audit in two years to review ADOT's progress in managing
construction staff size.
ADOTfS Construction Engineering Manpower Management System
Is Not Being Utilized Effectively And
Needs Greater Management Support ( see pages 27 through 36)
ADOT has not effectively used its construction engineering manpower management
system ( CEMMS) to adjust staffing levels based on workload. CEMMS was
implemented in October 1982 to provide construction managers with tools to plan,
schedule and control construction engineering personnel and costs.
We found, however, that despite the $ 550,000 ADOT has committed to CEMMS since
1982, CEM MS is not being effectively utilized to manage CE resources.
a ADOT has not effectively used CEM MS' long- range labor planning and
short- range project staffing functions to ensure that staffing levels are
appropriate.
0 The CEM MS budget function is not well understood or used by district personnel.
e ADOT has misused the CEM MS scheduling function, which if used properly,
assesses short- term staffing needs and plans for efficient personnel use. For
example, resident engineers have, in some cases, overstated staffing needs.
a CE M MS reports identifying A DOT construction engineering hours and cost
overruns are not received by nearly half of the construction orgs. Some districts
have stopped distribution of the reports. Without feedback, action cannot be
taken to control costs. District personnel stated that even when problem
projects are identified, l i t t l e or no formal action i s taken.
Although repeatedly informed of problems with CEMMS, ADOT management has
taken l i t t l e action to correct deficiencies or ensure system utilization. Problems
with management direction, adequate d i s t r i c t s t a f f commitment, system analyses and
updating and training have been reported to ADOT management regularly over the
last five years by outside consultants and ADOT staff. ADOT upper management
needs to make a f i r m commitment to address the problems w i t h i t s manpower
management system to ensure that ADOT uses i t s s t a f f and other resources
e f f i c i e n t l y and effectively.
TABLE OF CONTENTS
Page
INTRODUCTION AND BACKGROUND. . . . . . . . . . . . . . . . . . . . 1
FINDING I : ADOT HAS NOT ADEQUATELY CONTROLLED CONSTRUCTION
STAFFING LEVELS, RESULTING IN OVERSTAFFING THAT HAS
COST THE STATE AN ESTIMATED $ 7.4 MILLION OVER THE LAST
FOURFISCALYEARS . . . . . . . . . . . . . . . . . . 9
Analyses Show That ADOT Construction
Function I s o v e r s t a f f e d . . . . . . . . . . . . . . . . . . . 10
F i e l d V i s i t s And ADOT Analyses
Also Indicate O v e r s t a f f i n g . . . . . . . . . . . . . . . . . . 17
A l t e r n a t i v e s Should Be Considered
Before Increasing S t a f f Levels . . . . . . . . . . . . . . . . 19
The L e g i s l a t u r e Should Control
Construction S t a f f Levels . . . . . . . . . . . . . . . . . . 25
Recommendations. . . . . . . . . . . . . . . . . . . . . . . . 26
FINDING I I : ADOT'S CONSTRUCTION ENGINEERING MANPOWER
MANAGEMENT SYSTEM IS NOT BEING UTILIZED
EFFECTIVELY AND NEEDS GREATER MANAGEMENT SUPPORT . . . 27
CEMMS I s Not E f f e c t i v e l y Used To Manage CE Resources . . . . . 28
The System I s Not F u l l y Supported. . . . . . . . . . . . . . . 30
ADOT Needs To Take Action
To Make CEMMS Viable . . . . . . . . . . . . . . . . . . . . . 33
Recommendations . . . . . . . . . . . . . . . . . . . . . . . 35
OTHER PERTINENT INFORMATION. . . . . . . . . . . . . . . . . . . . 37
AREAS FOR FURTHER AUDIT WORK . . . . . . . . . . . . . . . . . . . 39
DEPARTMENT RESPONSE . . . . . . . . . . . . . . . . . . . . . . . 41
LIST OF TABLES
Page
TABLE 1 - Estimated Expenditures On
Highway Construction Program
Fiscal Years 1986- 87 Through 1990- 91 . . . . . . . . . . 7
TABLE 2 - Excess Construction S t a f f As Determined
By An Analysis Of Construction Engineering ( CE) Costs
For Fiscal Years 1983- 84 Through Three
Quartersof1986- 87 . . . . . . . . . . . . . . . . . . 11
TABLE 3 - Excess Construction S t a f f As Determined By
An Analysis Of Construction S t a f f P r o d u c t i v i t y
For Fiscal Years 1983- 84 Through Three
Q u a r t e r s O f 1 9 8 6 - 8 7 . . . . . . . . . . . . . . . . . . 12
TABLE 4 - D i s t r i c t and Statewide P r o d u c t i v i t y
Fiscal Years 1983- 84 Through
ThreeQuartersof1986- 87 . . . . . . . . . . . . . . . 13
TABLE 5 - Excess Construction S t a f f As Determined By An
Analysis Of Miscellaneous Time Charged For
Fiscal Years 1983- 84 Through Three
Quarters Of 1986- 87 . . . . . . . . . . . . . . . . . . 15
Table 6 - Costs Of Construction O v e r s t a f f i n g :
Fiscal Years 1983- 84 Through 1986- 87 . . . . . . . . . . 16
TABLE 7 - Estimated Construction S t a f f i n g Needs
Fiscal Years 1988 Through 1992 . . . . . . . . . . . . . 20
TABLE 8 - Percentage Of Contracting For Construction
Engineering Needed With Current 600 FTE S t a f f Level
For Fiscal Years 1987- 88 Through 1991- 92 . . . . . . . . 24
TABLE 9 - Construction Engineering Cost Percentage
Comparisons Of Completed Consultant And
ADOT Managed Construction Projects
Fiscal Years 1983- 84 Through 1986- 87 . . . . . . . . . . 38
LIST OF FIGURES
FIGURE 1 - ADOT Engineering D i s t r i c t s . . . . . . . . . . . . . . 3
FIGURE 2 - Highways D i v i s i o n Organizational Chart . . . . . . . . 4
FIGURE 3 - Relationship Of S t a f f i n g Levels To Workload:
Fiscal Years 1983- 84 Through Three
Q u a r t e r s O f 1 9 8 6 - 8 7 . . . . . . . . . . . . . . . . . . 22
INTRODUCTION AND BACKGROUND
The Office of the Auditor General has conducted a performance audit of the Arizona
Department of Transportation ( ADOT) in response to a July 26, 1985, resolution of
the Joint Legislative Oversight Committee. This performance audit was conducted
as part of the Sunset Review set forth in Arizona Revised Statutes 5541- 2351 through
41 - 2379.
This is the fifth in a series of reports to be issued on the Arizona Department of
Transportation. This report focuses on the management of Arizona's highway
construction program.
Construction Management Responsibilities
The Arizona Department of Transportation's Highways Division is responsible for the
construction of all interstate, primary and secondary roadways in the State's highway
system. There are two major groups within this Division: Highway Development and
Highway Operations. The Highway Development Group has responsibility for roadway
design and acquiring the land needed for construction of these roads. Highway
Operations, which is the focus of this audit, is responsible for coordination,
management and oversight of all construction to ensure that it meets specifications
and that contractors fulfill their contractual obligations.( Given the cost of
road construction, this function is critical to ensure that roads are of sufficient
quality to meet their intended design life.
Several specific activities comprise the Highway Operations Group's responsibilities.
Under the general category of construction engineering, these responsibilities include:
m Inspection and materials testing of all roadway construction to ensure that
procedures and materials meet plans and specifications.
( 1 ) A l l road construction i s performed by p r i v a t e contractors selected through a
competitive bidding process.
0 Surveying activities ( over and above those performed by the contractor)
necessary for effective control of construction operations.
0 Reviewing roadway design drawings.
o Taking measurements of estimated work completed for preparation of payments
made to contractors.
0 Reviewing, monitoring and processing claims, change orders and force accounts.
Most of these activities are the responsibility of ADOT's own personnel. However,
since fiscal year 1983- 84 ADOT has contracted with engineering consulting firms to
perform construction engineering activities for a limited number of construction
projects. While the number of these contracts has been minimal, the size of some
construction projects on which consultants have been utilized is large. During fiscal
year 1986- 87, consultants performed project oversight and inspection functions for
approximately 25 percent of the program dollars spent on actual road construction. a
Qrganization And Staffing
Within the Highway Operations Group, management of road construction projects is
the responsibility of the four engineering districts ( see Figure I), with technical
support provided by the Materials and Construction Sections located in AD 0 T's
central office. Figure 2 shows how ADOT is organized to carry out its construction
engineering responsibilities. A II construction related activities within these
engineering districts are the responsibility of the district engineer. Actual
day- to- day operations are delegated to the deputy district engineer.
FIGURE 1
ADOT ENGINEERING DISTRICTS
Offices
FIGURE 2
HIGHWAYS DIVISION
ORGANIZATIONAL CHART
Assistant Director
State Engineer
Chief Deputy
State Engineer 1
District
Engineer
Deputy District
Engineer
2 - -
Area Engineers
Construction
Source: ADOT Construction Manual.
4
Because of the large geographical areas encompassed, engineering districts are
further divided into three or more area jurisdictions. Each of these is headed by an
area engineer. Area engineers are responsible for all highway construction within
their respective geographical boundaries.
Construction personnel are assigned to construction or materials testing units ( orgs).
ADOT's construction orgs, under the supervision of a resident engineer, are
responsible for the day- to- day field coordination and management ( construction
engineering) of various construction projects assigned to them.
ADOT's Materials and Construction Sections in the central office provide the
districts with technical support in fulfillment of these construction management
responsibilities. The Materials Section develops test methods, acts as the reference
laboratory for district and project laboratories, and performs tests that cannot be
conducted by the other labs. The Materials Section also monitors tests performed by
project and district labs to ensure uniform test results. The Construction Section is
responsible for quality control site inspections and maintaining the Division's
Construction Engineering Manpower Management System. This Section also audits
and processes contractor payments and manages all paperwork pertaining to
contractor claims.
The majority of ADOT personnel involved in the coordination and management of
highway construction projects are located in the four engineering districts. There are
approximately 600 permanent construction full- ti me equivalent ( FTE) positions
assigned to these districts. Another 128 FTEs are assigned to the Highways Division
and Highway Operations central office administration, and to the Materials and
Construction Sections.
Revenues And Expenditures
Funds for Arizona's highway construction program come from a variety of Federal
and State sources, including gasoline and sales taxes. The majority of Arizona's share
of Federal highway funds is earmarked for completion, improvement and preservation
of the interstate highway system.
Federal funds are still a major funding source for the highway construction program,
expected to total $ 644 million over the next five years. However, the Federal share
has declined significantly in recent years and represents just 22 percent of the funds
allocated to the program from fiscal year 1987- 88 through fiscal year 1991- 92. This
contrasts significantly with prior years when the Federal share averaged
approximately 67 percent of total funds allocated.
Approximately 77 percent ($ 2.27 billion) of the highway construction program over
the next five years will be financed with State revenues. This represents a major
increase over previous years.
Increased State funding is the result of legislation ( House Bill 2306) enacted in 1985.
This legislation increased fuel taxes by 3 cents per gallon, effective January 1, 1986,
and reallocated a share of vehicle license taxes for highway construction. - a
Furthermore, House Bill 2306 gave each county the option of having its voters
approve a one- half cent increase in sales taxes, revenues from which would be
earmarked for construction of county controlled- access roadways.
A one- half cent increase in sales taxes ( Proposition 300) was approved by Maricopa
County voters in October 1985. The sales tax increase became effective as of
January 1, 1986, with proceeds going to highway construction.
Bond revenues represent an important vehicle for accelerated financing of highway
construction in urban and rural areas of the State. Nearly $ 1 billion in bonds have or
will be issued by AD OT between fiscal years 1987 and 1991, with $ 850 million funding
controlled- access routes in Maricopa County and $ 16 million in Pima County. The
balance of the bond proceeds will finance Statewide highway improvements.
These revenue sources not only cover the cost of land acquisition and roadway design
and construction, they also fund A DOT'S construction oversight costs. If A DOT can
realize greater efficiencies in its oversight responsibilities, more funds would be
made available for development of Arizona's highway system.
Expenditures for purchase of land and design and actual construction of new highways
during fiscal year 1987- 88 will total approximately $ 716 million. This contrasts with
fiscal year 1985- 86 and fiscal year 1986- 87 expenditures of approximately $ 371
million and $ 578 million, respectively. Over the next five years expenditures will
total approximately $ 2.93 billion ( see Table 1). This sharp increase in highway
program expenditures is largely due to the initiation of an ambitious program to
expand Maricopa County's urban freeway system.
TABLE 1
ESTIMATED EXPENDITURES ON HIGHWAY CONSTRUCTION PROGRAM
( F I SCAL YEARS 1986- 87 THROUGH 1990- 91 ) ( a)
Fiscal Year
Total Estimated
Estimated Construct ion
Expendi tures ( b) Expend i tures
( a) Funds released during any specific fiscal year w i l l not necessarily be
expended during the same f i s c a l year. For example, a one year project
authorized at the end of a given fiscal year would not be completed u n t i l
the following fiscal year.
( b) These cost figures include estimated land acquisition ( right- of- way),
highway design and roadway construction ( including construction
engi neeri ng) expendi tures.
( c) The cost figures do not include local government projects overseen by ADO1
which are estimated to total $ 25 m i l l i o n annually.
Source: Arizona Department of Transportation's Five- Year Transportation
Facilities Construction Program Report ( fiscal year 1987- 88).
Audit Scope and Purpose
Our audit focused on the staffing levels required by the Arizona Department of
Transportation's Highway Operations Group to effectively and efficiently manage the
State's highway construction program. The report presents findings and
recommendations in the following areas.
@ A DOT'S construction staffing levels.
( I A B OTIS Construction Engineering Manpower Management System ( CE M MS).
In addition, we developed other pertinent information concerning contracting for
construction oversight. ( see pages 37 through 38)
Due to time constraints, we were unable to address all potential issues identified
during our audit. The section Areas For Further Audit Work ( see pages 39 through 40)
describes these potential issues.
Our audit was conducted in accordance with generally accepted governmental
auditing standards.
The Auditor General and staff express appreciation to the Director and staff of the
Arizona Department of Transportation for their cooperation and assistance during the
course of our audit.
FINDING I
ADOT HAS NOT ADEQUATELY CONTROLLED CONSTRUCTION STAFFING LEVELS
RESULTING IN OVERSTAFFING THAT HAS COST THE STATE
AN ESTIMATED $ 7.4 MILLION OVER THE LAST FOUR FISCAL YEARS
The Arizona Department of Transportation ( ADOT) has not adequately controlled
construction staffing levels. Our analysis of construction engineering costs and other
Statewide data indicates that overstaffing has cost the state an estimated $ 7.4
million over the last four fiscal years. Field visits and ADOT's own reviews further
indicate that ADOT has been overstaffed. Before increasing staff levels to meet
short- term program increases, A D 0 T should first consider making greater use of the
alternatives it has to match staffing levels and workload. To ensure that ADOT
staffs its construction engineering program appropriately, the Legislature should
annually set and monitor construction staff levels.
Methodology
We used three different quantitative methods to determine whether A D 0 T's
construction engineering program was appropriately staffed. Using different
analytical methods helped to validate results. The analytical methods used were
based on ADOT construction engineering cost goals, ADOT's past productivity and
A DOT miscellaneous time ( non construction related hours charged by A DOT staff).
In addition to analyzing several types of quantitative data, we visited district, area
and construction orgs in the four districts to discuss and observe staffing needs. We
also reviewed ADOT reports that studied staffing needs. Both the field visits and the
ADOT reports were used to determine whether results obtained through analyses of
Statewide data were reasonable.
Analyses Show That A DOT
Construction Function Is Overstaffed
Results of all quantitative analyses we conducted indicate A DOT'S construction
function has been overstaffed the last several years. '" Analyses of
construction engineering costs, productivity and non construction related time
charged by staff all indicate that ADOT has had more construction staff than the
work required. Overstaffing over the last four years has cost the State an estimated
$ 7.4 million.
Construction Engineering costs analysis indicates overstaffing - Analysis of
construction engineering ( CE) '*' costs indicates overstaffing of 21 to 101 staff
from fiscal years 1983- 84 through 1986- 87. This analysis used A DOT's own CE cost
goals as criteria in determining appropriate staffing for these years. These goals,
expressed as a percentage of total project costs, were considered reasonable
because they fall in the mid- range of other states' CE costs. ( 3) Several other
states reported lower CE goals and costs. Further, when compared to historically
attained productivity ( see productivity analysis on page 111, A DOT could meet the
CE cost goals without working at peak productivity.
The analysis determined the amount of staff needed per year using A DOT's CE cost
goals as criteria. To determine needed staff, the appropriate CE cost goal was
applied to total payments to contractors for each year to determine the allowed
oversight cost amount. This amount was then divided by the historical construction
staff- related cost per labor hour for the appropriate year to determine the amount
ADOT maintains it has s t e a d i l y reduced i t s number of construction s t a f f since 1976.
ADOT prepared a schedule showing the r a t i o of s t a f f t o construction d o l l a r s has also
decreased. We d i d not i n c l u d e t h i s schedule i n our r e p o r t because the s t a f f i n g data
shown i n the schedule cannot be v e r i f i e d . Furthermore, ADOT8s analysis would not
a f f e c t r e s u l t s o f our c u r r e n t analysis which shows o v e r s t a f f i n g has occurred t h e
past f o u r years.
( 2) CE costs are a l l p r o j e c t o v e r s i g h t ( i n s p e c t i o n , survey, l a b o r a t o r y and
a d m i n i s t r a t i o n ) r e l a t e d costs borne by ADOT from the time a p r o j e c t s t a r t s through
completion. H i s t o r i c a l l y , 1 abor expenses have comprised from 75- 90 percent o f t o t a l
CE cost. CE cost i s expressed as a percentage of the cost of c o n s t r u c t i o n ( what i s
paid t o the c o n t r a c t o r a c t u a l 1 y b u i l d i n g t h e r o a d ) . ADOT's c o n s t r u c t i o n
s t a f f - r e l a t e d CE cost goals vary depending upon the size of the p r o j e c t , and range
from 6.3 t o 19.8 percent. The l a r g e r the p r o j e c t , the lower the CE percentage
should be and v i c e versa.
( 3 ) Of the 46 s t a t e s surveyed, 25 s t a t e s responded t o questions concerning CE costs.
of oversight hours needed. Oversight hours needed were then divided by an hours per
labor year figure based on historical data to calculate the total staff needed for the
year. Overstaffing amounts were then determined by subtracting staffing needs from
actual staff size. Overstaffing, as determined by CE cost analysis, is shown in Table
2.
TABLE 2
EXCESS UHSTRUCTIOW STAFF AS DETERWINED BY AN ANALYSIS
OF COWSTRUCTIOW ENGINEERING ( CE) COSTS
FOR FISCAL YEARS 1983- 84 THROUGH THREE QUARTERS OF 1986- 87
CE Cost Hours
Fiscal Contractor ( a) CE ( b) Cost Per( c) Hours Per ( dl Staff( e) Actual( e1 Excess
Year Payaent s x Goal = GAo i Hour = Needed i Year = Needed vs. Staff Staff
1983- 84 $ 165.481.277.30 x 8.70% $ 14.396.871.13 i $ 16.18 = 889.794 t 1.789 = 497 vs. 544 - 47
1984- 85 202,097,225.12 x 8.14% = 16.450.714.12 + 17.21 = 955.881 i 1.789 = 534 VS. 591 57
1985- 86 184.652.465.89 x 8.542 = 15.769.320.59 + 17.17 = 918.422 + 1.789 = 513 VS. 614 101
1986- 87 179.807.627.33 x 7.85% = 14.114.898.75 i 18.63 = 757.643 + 1.342 - 565 VS. 586 21
( a) Excludes projects administered by private consulting firms.
( b) Overall goal based on cumulative application of ADOT CE cost goals minus h i s t o r i c a l non construction org CE costs.
On the average, 79 percent of a l l CE costs were charged from construction orgs.
( c) Based on AWT1s construction s t a f f related costs and construction hours worked.
( d) Average nrwber of construction engineering hours worked including overtime per year by one eqloyee.
( e) Does not include staff needed t o oversee consultant unaged projects.
Low productivity 1- Low produc tivity by ADOT
construction staff over the last four fiscal years also indicates overstaffing. Our
analysis indicated overstaffing ranging from 38 to 207 staff when ADOT was held to
historically attainable productivity standards.
The productivity analysis, like the CE cost analysis, examined the efficiency with
which staff oversee and process construction projects. The analysis compared how
much construction work ( payments to contractors in dollars) was processed per ADOT
staff hour charged to construction oversight. A historically attainable productivity
indicator ($ 163 per hour) was then selected as criterion for determining how many
staff were needed per fiscal year i f staff worked at that productivity level. To
calculate ideal staff size, first, payments to contractors for the year ( the amount of
work to be processed by ADOT staff) were divided by the productivity standard to
determine total hours needed to process the work i f working at the productivity
standard. Then, the total hours needed were divided by 1,789 ( historical average
number of construction related hours charged by one employee for one year) to
calculate needed staff. Needed staff were then subtracted from actual yearly staff
totals to determine overstaffing. Overstaffing as shown by productivity analysis is
shown in Table 3.
TABLE 3
EXCESS CONSTRUCTION STAFF AS DETERMINED BY AN ANALYSIS
OF CONSTRUCTION STAFF PROWCTIVITY
FOR FISCAL YEARS 1983- 84 THROUGH THREE OUARTERS OF 1986- 87
Adjusted ( a) Average
Fiscal Contractor Productivity Hours Yearly Staff ( b) Actual ( b) Excess
Year Payments .?- Standard = Needed + StaffHours = needed vs. Staff =
1983- 84 $ 140,203,179.43 4 $ 163.17 = 859,246 + 1789 -- 480 V S . 544 = 64
1984- 85 149,831,650.41 s 163.17 = 918,254 + 1789 - 513 VS. 591 = 78
1985- 86 118,949,030.21 + 163.17 = 728,988 i 1789 -- 407 V S . 614 = 207
1986- 87 119,891,185.03 + 163.17 = 734,762 t 1342 -- 548 VS. 586 = 38
( 3 Quarters)
.,., c,,,, rrn. ,,.,,,,, & , , A : err.. l,. n, rU,,,,,,, Staff a- a! y: l: cf A!?!?? Csn: trcct! c:: Cczt: an6 Lahcr k ' t a .
( a) Amounts adjusted for i n f l a t i o n using Arizona Construction Cost Index.
( b) Does not include s t a f f needed to oversee consultant managed projects.
$ 163 per hour was selected as the productivity standard for several reasons. First,
the standard has already been achieved by ADOT on a Statewide basis in fiscal year
1986- 87. Second, the standard has been achieved or exceeded by all districts except
District Two in at least two of the last four years. The standard represents from 77
to 92 percent of the peak productivity of all the districts except
District Two. Third, because District Two's productivity lowers the standard, the
$ 163/ hour figure is actually somewhat conservative. ( ' I District and Statewide
productivity over the last four fiscal years is shown in Table 4.
TABLE 4
DISTRICT AND STATEWIDE PRODUCTIVITY
FISCAL YEARS 1983- 84 THROUGH THREE QUARTERS OF 1986- 87
D i s t r i c t D i s t r i c t D i s t r i c t D i s t r i c t
Year 1 2 3 4 ( a) statewideib)
1986- 87 178 104 171 196 163
( Three Quarters)
( a) While D i s t r i c t Four yearly p r o d u c t i v i t y figures were on the average higher than
other d i s t r i c t s , i t also had the widest variance i n q u a r t e r l y p r o d u c t i v i t y
rates. For example, i n f i s c a l year 1985- 86, D i s t r i c t Four p r o d u c t i v i t y varied
from $ 80/ hour t o $ 241/ hour. This variance i s due to seasonal changes i n
D i s t r i c t Four workload with lowest p r o d u c t i v i t y o c c u r r i n g d u r i n g w i n t e r months.
( b) Statewide figures d i f f e r from an average of the four d i s t r i c t s because of
varying workloads per d i s t r i c t .
Source: Auditor General analysis of ADOT construction costs and labor data.
According to current and former ADOT s t a f f , D i s t r i c t Two has been overstaffed i n
past years, p a r t i a l l y because of an expected increase i n construction a c t i v i t y due
t o a sales tax i n i t i a t i v e t h a t subsequently d i d n ' t pass. Because D i s t r i c t Two has
q u a l i t i e s s i m i l a r t o a l l o f the other d i s t r i c t s ( e i t h e r a large urban area, o r r u r a l
and mountainous areas), we determined it was reasonable to hold D i s t r i c t Two t o
s i m i l a r p r o d u c t i v i t y expectations.
Excessive miscellaneous time charged - Even a limited analysis using
miscellaneous ( non construction related) ti me charged by staff indicates that A D 0 T
has had more construction staff than the work required.
Construction staff charge time to a ' lmiscellaneousll code when performing non
construction related tasks. Miscellaneous activities can occur because they are
scheduled such as training sessions, or because of lack of construction work because
of funding, weather, contractor problems or project scheduling changes. Auditor
General Staff field visits to construction orgs found that construction staff perform
a variety of tasks on miscellaneous time. These include, for example, road
maintenance, building maintenance and painting. One org constructed a loading
dock and put skirting around a trailer.
Unlike CE and productivity analyses, miscellaneous ti me analysis does not examine -
efficiency. The analysis only identifies excessive miscellaneous hours charged, and
does not identify whether excessive staff are assigned to construction jobs, or
whether they work at an adequate pace. ( ' ) Our analysis revealed that A D 0 T's
excessive miscellaneous time over the past four fiscal years translated into
overstaffing ranging from 17 to 32 staff, except for fiscal year 1984- 85 when AD 0 T
staff did not use ail of the miscellaneous time allowed. Our analysis simply
caiciilaied the excessive amounts of misce!! zneous time ( ameunts ever the P. Ll O?
allowed 10 percent of total time) using ADOT labor data, and converted that to
excess staff by dividing total excess hours by the average number of hours worked
per year per employee. Excess staff as determined by an analysis of miscellaneous
time is shown in Table 5 ( see page 15).
( ' I AM) T analyses discussed on Page 17 suggest t h a t a t times ADOT has not worked
e f f i c i e n t l y . Excessive time has been charged on some occasions, however, t h e extent
of t h i s problem i s unknown.
Miscellaneous time was excessive in several construction orgs analyzed. AD 0 T
management stated that miscellaneous time should not exceed 10 percent of total
time charged by staff. Our analysis, however, found some orgs with more than 60
percent miscellaneous t i m e f o r some months. From January 1985 through March
1987, 10 of the 32 construction orgs averaged from 20 to 36 percent miscellaneous
time.
TABLE 5
EXCESS COnSTRUCTION STAFF AS DETERMINED BY AN ANALYSIS
OF MISCELLANEOUS TIME CHARGED
FOR FISCAL YEARS 1983- 84 THRWaf THREE WARTERS OF 1986- 87
Total 102 A1 lowed Actual Average
Fiscal Hours Mi scel 1 aneous Mi sc. ~ i s c . Year1 y Excess
Year Charged X Allowance = Time vs. Time = Difference + S t a f f H o u r s = Staff
1983- 84 1,087,309 X .10 = 108,731 V S . 166,058 = 57,327 + 1789 = 32
1985- 86 1,256,224 X .10 = 125,662 VS. 177,010 = 51,388 + 1789 = 29
1986- 87 889,526 X .10 = 88,953 VS. 111,866 = 22,913 + 1342 = 17
( Three Quarters)
Source: Auditor General S t a f f analysis o f ADOT Construction Costs and Labor Data
( a) Statewide, ANT s t a f f accrued less than the 10 percent allowable miscellaneous time; therefore, the
analysis was not applicable.
Overstaffing has cost the State $ 7.4 million - Overstaffing over the last four
fiscal years has cost the State approximately $ 7.4 million. The cost of overstaffing
was determined by multiplying the number of excess staff ( as determined by the CE
cost analysis) by the average cost per employee per year. We selected the CE cost
analysis as our primary indicator of overstaffing because it was generally more
conservative than the productivity analysis. We felt this gave A DOT the " benefit of
the doubt" on differences between the analyses. Costs of construction overstaffing
over the last four fiscal years are shown in Table 6.
TABLE 6
COST OF CONSTRUCT l ON OVERSTAFF I NG :
F l SCAL YEARS 1983- 84 THROUGH 1986- 87
F l SCAL EXCESS AVERAGE STAFF OVERSTAFFING
YEAR STAFF ( a) WAGE COST ( b) COST
TOTAL $ 7,443.064
( a) Based on Construction Engineering Cost Analysis.
( b) Based on h i s t o r i c a l construction- related costs including equipment and travel costs
incurred while working on construction projects, plus a 10% miscellaneous time cost
a1 1 owance.
Source: Auditor General Staff analysis of ADOT construction costs and labor
data.
Field Visits And ADOT Analyses
Also Indicate Overstaffing
Field visits and A DOT analyses support the conclusion that A DOT has overstaffed its
construction function. Field visits to the engineering districts provided several
examples of overstaffing and under utilization. In addition, AD OT field reviews have
demonstrated overstaffing.
Overstaffing seen during field visits - Field visits by Auditor General staff to the
four engineering districts provided examples of overstaffing and staff under
utilization in at least two districts. We visited most construction orgs in the State,
and spoke with district and area management and staff. The following examples
illustrate staffing inefficiencies.
e The District Two engineer acknowledged that his district is currently overstaffed
by 50 to 60 full- time equivalent employees ( FTEs). The district expects to have
an average surplus staff for the next 12 months, of approximately 44 FTEs.
Additionally, the engineer indicated that he would not become concerned about
district staff size until it was reduced to 100 FTEs ( current staff size is
approximately 140 FTEs). According to other current and former ADOT staff,
this district has been overstaffed in past years. The district had expected
increased construction that did not occur.
e District Four maintains a nearly full construction work force ( averaging
approximately 70 staff) throughout the winter, even though construction work is
primarily conducted from May through October due to climatic conditions in this
district. Although some construction related tasks are performed, most staff are
involved in non construction duties such as maintenance, snowplowing assistance,
paperwork and training. One org resident engineer said that his staff ran out of
things to do this past winter.
e District Four did not appear to be utilizing temporary staff and summer interns
efficiently. The district normally hires approximately 30 temporary staff and
summer interns each spring to help the large amount of work that is done during
the peak construction season. However, our visits to the two orgs that have the
temporary staff indicated that there was no construction related work for them
and that there would not be much for a few more months. To keep the
temporaries busy, some were assigned to " double- up" with regular ADOT staff.
Org staff have indicated that by the time the " crunchu occurs, the temporaries
and interns are not available because they have returned to school.
a One resident engineer in District Four stated that he had surplus staff for most
of the year. Currently, his org is staffed with approximately 20 permanent
employees year- round and approximately ten temporary staff during the
summer. However, because workload is seasonal, staffing requirements vary.
He stated that the org needed six workers for three months, 12 workers for six
months, and 30 workers for three months. As a result, the org is essentially
overstaffed for three- quarters of the year.
ADOT analyses also show overstaffing - ADOT field reviews also demonstrate
overstaffing. ADOT Field Review Services conducts reviews of orgs or projects,
reporting on several construction- related items, including staff utilization.
Overstaffing found in field reviews are shown in the following examples.
A recent ADOT review revealed surplus staff in two orgs i n District Four. One
org, with an average staff size of 19, had a surplus of 12 staff ( approximately 63
percent surplus). The other org, with an average staff size of 28, had a surplus of
ten staff ( approximately 36 percent surplus). A DOT figures calculate the excess II
labor cost for those two orgs for the two months examined at more than -
$ 102,000. Auditor General field visits to the same two orgs found similar staff
under utilization.
a A 1984 ADOT field review found excessive survey hours charged by ADOT crews
on a particular project in District Two. A comparison of estimated time needed
to perform the work vs. actual hours charged showed that an excess of 6,000
hours was charged. Those excess hours translate into nearly five surplus staff to
perform survey activities for the duration of the project ( 280 working days). In
addition, according to central office staff, another org in the district had excess
e,, v,, m, t e+ aFF and t: ansfe:: ed them te the erg assigfied t~ this praject. aul * by * L o 1 l
a Another 1984 field review of two projects in District One revealed excessive
amounts of time charged by ADOT survey crews for two work activities. One
construction activity overran ADOT estimates by 241 percent. The other
activity had an overrun of 315 percent. Central office also questioned the use of
two survey crews, when only one was actually needed, because the two projects
were adjacently located. The excessive hours charged to survey ( in contrast to
planning estimates) could possibly be related to the extra survey crew assigned
to the project. The number of staff assigned to office duties was also
questioned. There were four employees in the office, when actually only one was
needed. Furthermore, the number of supervisory and lab staff on another project
were considered to be excessive.
a ADOT recently reviewed a project in District Three and found a significant
overrun of hours charged to the preliminary phase of the project. Based on
ADOT estimates, ADOT crews overran hours needed for this phase of the job by
251 percent. According to Central Office staff, the overrun can be attributed to
transfers of staff from other districts. In addition, further analysis by Auditor
General staff indicated that hours charged by District Three staff were
approximately the number of hours needed to perform the function. Therefore,
excessive hours charged by staff from the other districts may have contributed
to the significant overrun.
Alternatives Should Be Considered
Before Increasing Staff Levels
Before adding construction staff to meet short- term increases in workload, ADOT
should first consider using alternative methods for providing construction oversight.
Staff needs will flutuate both over and under current levels over the next five years.
A DOT may not have to increase staff to meet short- term program increases i f A DOT
increases its use of consultant engineers and exercises other options for adjusting
workload for two out of the next five years.
Staff needs will fluctuate over the next five years - As shown in Table 7, page 20,
ADOT's construction staffing needs will fluctuate over the next five years. Based on
an analysis using CE cost goals,"' ADOT's staff needs will increase significantly
for two of the next five years. However, at current levels ADOT will be overstaffed
in two of the other three years.
( 1) The CE cost analysis i s the same as the method described on page 10, except f u t u r e
construction estimates from ADOT's f i v e year statewide and l o c a l government plans
were used. Again, t h i s represents a more conservative estimate than the estimates
based on ADOT's h i s t o r i c a l l y a t t a i n e d p r o d u c t i v i t y .
TABLE 7
ESTIMATED CONSTRUCTION STAFFING NEEDS
FISCAL YEARS 1988 THROUGH 1992
DIFFERENCE
F l SCAL CURRENT PROJECTED ( current minus
YEAR STAFF ( a) STAFF NEEDS ( b) projected)
( a) For 1986- 87, ADOT's c o n s t r u c t i o n work force averaged 600 FTEs, and i s expected t o
remain a t or above t h a t l e v e l .
( b) Future s t a f f p r o j e c t i o n s were based on the f o l l o w i n g assumptions. F i r s t , ADOT would
continue t o contract out approximately 25 percent of i t s construction f u n c t i o n t o
p r i v a t e f i r m s or l o c a l governments ( i n f i s c a l year 1986- 87, ADOT contracted out 25
percent t o p r i v a t e f i rms and l o c a l governments). Second, ADOT1s c o n s t r u c t i on- related
CE cost goal would be 7.85 percent ( t h e same as i n 1986- 87), and ADOT would be able
t o meet t h a t goal. F u r t h e r , c o n s t r u c t i o n costs were assumed t o increase by 5
percent. We projected c o n s t r u c t i o n workload based on the f i v e year statewide and
l o c a l government construction plans estimates according t o a formula provided to us
by ADOT Administrative Services D i v i s i o n .
Source: Auditor General analysis of ADOT f i v e year construction plans.
ADOT could manage workload fluctuation more efficiently - ADOT could manage
its workload more efficiently i f it made better use of alternatives to increased
staffing levels. Our analyses indicate that A DOT has historically maintained a high
base staffing level although workload fluctuates in seasonal and multi- year cycles.
As shown in Figure 3 ( see page 22), staffing has remained relatively stable while
workload has experienced both highs and lows. Although costly and inefficient,
maintaining a high permanent staffing base has enabled ADOT to meet much of its
peak workload Statewide with existing staff. ADOT has made some use of
alternative approaches to managing peak workloads. However, it could operate more
efficiently i f these approaches were used more widely and effectively. These
alternative approaches include:
a Use of consultant engineers - Currently, ADOT contracts approximately 25
percent of i t s total work to consultant engineers, who perform construction
engineering functions normally done by - A D O T staff. ADOT has used
consultants on technically complex projects, such as the 1- 1011- 17 interchange
( the stack), and on projects ADOT did not want to oversee for other reasons. A
primary benefit of using private consultants, however, is that ADOT only has to
pay for this service when it is needed, thus reducing non productive time.
ADOT could make greater use of consultants to manage fluctuations in
workload. Consultants could be used, for example in District Two which
expects a temporary increase in its workload over the next few years.
a Contract out functions - ADOT could increase its contracting out of specific
construction engineering functions. Consultant engineers subcontract for
survey crews and laboratory testing functions as needed, thus reducing non
productive time. ADOT could adopt the same approach. According to a survey
of other states, many states contract for surveying, lab and inspection
functions. ( 1 ) According to the State of Colorado, having contractors
perform staking lowered overall CE costs from 11 to 9 percent.
( ' 1 Auditor General s t a f f compiled a survey regarding s t a t e highway construction and
d i s t r i b u t e d i t to each transportation department across the country. Of the 50 surveys
mailed, 46 states responded.
FIGURE 3
RELAT l ONSH l P OF STAFF l NG LEVELS TO WORKLOAD : ( a)
F I SCAL YEARS 1983- 84 THROUGH THREE QUARTERS
OF 1986- 87
July 1983 July 1984 July 1985 July 1986
( a) Construction p r o j e c t s monitored by engineering consultants have been excluded from
t h i s analysis.
Source: Auditor General Staff analysis of ADOT Construction Staff and
contractor payments from July 1983 to March 1987.
a Use of temporary staff - ADOT could also increase and make more timely
use of temporary staff to assist in meeting workload peaks. For example,
ADOT currently uses some temporary staff in District Four, although these
staff have not always been available when needed due to project scheduling
delays ( see page 17).( 1) Colorado makes extensive use of temporary staff
to meet its summer workload demands. Forty- one of the states we surveyed
indicated they use temporary employees to meet workload peaks.
a District transfers - ADOT could also make more extensive use of transfers
between districts to accommodate fluctuations in workload. As discussed
previously, staff in seasonal districts accumulate excessive amounts of non
productive time during the winter months. Transferring these staff or others in
districts experiencing reductions in workload would help ensure that staff
resources are used efficiently. For example, District II, from which staff have
been transferred in the past, will s t i l l be overstaffed in fiscal year 1987- 88 by
an average of 44 positions. Thirty- seven states responding to our survey
indicated they transfer staff among districts to handle workload shortages.
r Share resources with maintenance - ADOT may be able to utilize
maintenance staff to assist during peak workload periods. Twenty states
surveyed indicated that their construction and maintenance work forces
occasionally share labor resources. Currently, A D 0 T construction staff assist
maintenance crews during periods of low construction activity, however,
maintenance, which has approximately 780 staff statewide, does not assist in
construction during peak construction work periods.
Construction staff increases may be unnecessary - Before increasing staff size,
ADOT should first consider more extensive use of the above alternatives as a means
of handling workload fluctuations. We determined the portion of its program A DOT
would have to contract out over the next five years i f it maintained its current
staffing level of 600. As shown in Table 8 ( see page 241, a significant increase over
the current level of 25 percent usage would be required for only two of the five fiscal
years. The highest increase would be in fiscal year 1989- 90 when consultant usage
would total 33 percent. In two of the future fiscal years, consultant usage would
actually decline over current levels.
Analysis of D i s t r i c t Four s t a f f size revealed t h a t the permanent construction work
force average 70 FTEs f o r t h e p a s t two f i s c a l years, yet ADOT central o f f i c e
i n d i c a t e d the D i s t r i c t only needs a year- round s t a f f o f 45. This has r e s u l t e d i n
considerable non productive time during l o w c o n s t r u c t i o n periods. From November 1986
through A p r i l 1987, average miscellaneous time ranged from 15 t o almost 50 percent.
One org charged over 70 percent of i t s time t o the miscellaneous code f o r two
consecutive months.
TABLE 8
PERCENTAGE OF CONTRACT l NG FOR CONSTRUCT l ON ENG l NEER l NG
NEEDED WlTH CURRENT 600 FTE STAFF LEVEL
FOR FISCAL YEARS 1987- 88 THROUGH 1991- 92
F I SCAL CURRENT PERCENTAGE OF CONTRACTED'~! D I FFERENCE OVER/
YEAR CE PERCENTAGE CE NEEDED WITH 600 FTE UNDER CURRENT LEVEL
( a) AWT for FY 1986- 87 contracted out 25% of i t s total workload. ADOT expects to
contract out a s i m i l a r amount i n FY 1987- 88. a
Source: A u d i t o r General analysis o f ADOT c o n s t r u c t i o n c o s t s and l a b o r
data.
To increase control over staffing, the Legislature should set construction staff
levels. ADOT has failed to adequately adjust i t s staff levels in the past, even though
it developed an automated system to determine appropriate staffing. As is done in
many other states, the Legislature, through its appropriations process, could set and
monitor staffing levels.
Construction staff not adjusted based on workload - Our analysis revealed that
ADOT construction staff levels have not been adjusted based on workload. During
low periods in highway construction, ADOT has not decreased its staff levels.
ADOT implemented a manpower management system to assist in adjusting
construction staffing to meet workload fluctuations. ADOTts Construction
Engineering Manpower Management System ( CEMMS), however, has not been fully
utilized to assist in adjusting staff levels. ADOT management has not fully
implemented the system nor required full utilization by field staff ( see Finding II,
page 27).
Increased legislative involvement - Because A D 0 T has not adequately controlled
its staffing, the Legislature should consider setting and monitoring construction staff
levels. Currently, ADOT determines its construction staffing needs based on its
highway construction program. Neither the Legislature nor the Executive Budget
Office ( EBO) conducts a comprehensive review or regulates construction staff levels.
According to an ADOT official, it would not be feasible to legislate construction
staff levels, because workload cannot be accurately estimated.
However, results of a survey revealed that several other states do have their
construction staffing positions determined by either a legislative review committee
or an executive budget agency. Of the 46 states that responded to the survey, 18
indicated that a legislative review committee is involved in determining highway
construction staff positions. In addition, 13 states indicated that an executive budget
agency is directly involved in this function. Therefore, based on methods utilized by
other states to determine staffing levels, i t may be feasible for the Arizona
Legislature to consider setting and monitoring construction staff.
The Legislature could set staffing levels based on construction information provided
by ADOT. Several types of information could be used for monitoring and
appropriations decisions. These include reports of CE costs, construction staff
productivity, construction staff charges to miscellaneous time, future construction
workload ( estimates of future contractor payments), and the percentage of projects -
that A DOT expects to contract out. A DOT currently generates some of these reports
from the CEMMS system and could generate others, as was done by Auditor General
staff. Monitoring these reports on a regular basis would allow the Legislature to
determine i f ADOT is utilizing its staff effectively. Based on this information, the
Legislature could then appropriate staff levels accordingly.
RECOMMENDATIONS
1. ADOT should first consider increasing its use of alternative staffing methods as
needed to handle peak construction periods Statewide and in seasonal districts
before adding additional permanent staff.
2. The Legislature should consider setting and monitoring ADOT construction staff
levels.
3. The Legislature should consider requiring a follow- up performance audit in two
years of AD OTts construction management function to review its progress in
managing staff size.
FINDING II
ADOT'S CONSTRUCTION ENGINEERING MANPOWER MANAGEMENT SYSTEM
IS NOT BEING UTILIZED EFFECTIVELY AND NEEDS
GREATER MANAGEMENT SUPPORT
Although the Arizona Department of Transportation ( ADOT) has committed at least
$ 550,000 to its Construction Engineering Manpower Management System ( CEMMS)
since 1982, the system is not being utilized effectively and lacks adequate support.
CEMMS is not being used to effectively manage ADOT's construction engineering
( CE) resources. ADOT needs to more fully support the system and to address system
deficiencies if CEM MS is to become a viable management tool.
BACKGROUND
CEMMS was developed to assist ADOT in controlling its construction engineering
resources. In the early 1980s, ADOT recognized the need f o r b e t t e r methods of
projecting manpower needs and improved management of construction engineering
resources. In addition, although ADOT has approximately 600 construction
engineering employees, this work force is not appropriated by the Legislature.
Instead, ADOT determines work force size to fit its construction work program. As a
result of ADOT's need for improved management o f i t s CE resources and Legislative
concern over effective control of ADOT's work force, the Department developed a
construction management system.
CEM MS was designed to provide construction managers with tools to plan, schedule
and control construction engineering personnel and costs. The primary objective of
the design and implementation of CEM MS was to develop a system that would:
a Reduce construction engineering costs.
a Improve work and budgeting methods.
a Provide end products to determine the most effective use of ADOT construction
engineering personnel and equipment.
Development of CEMMS began in December 1981, and the system was implemented
Statewide on October 1, 1982. Since CEMMS was implemented, ADOT has
committed at least $ 550,000 to the system, of which an estimated $ 460,000 has been
incurred by central office staff, with an additional $ 93,000 paid to consultants. ( 1 )
CEM MS Is Not Effectively Used
To Manage CE Resources
CEMMS is not being utilized effectively to manage ADOT's CE resources. ADOT has
not effectively used CEMMS functions to plan and staff, budget, schedule, and
monitor these resources.
Planning and staffing functions not utilized effectively - CEM MS is designed for
short- term and long- term planning. Short- term planning is needed to determine the
number of employees required and which months they are needed, so plans may be
made to meet those needs. Long- term planning allows management to take action to
prevent significant staffing shortages or surpluses. ( 2)
However, the CEM MS planning and staffing functions are not being utilized
effectively to plan A DOTfs labor requirements. A DOT Highway Operations
management indicated that CEM MS was only slightly effective in assisting them to
forecast and control labor needs. District engineers stated that " CEMMS as a
manpower projection tool has been nearly useless." As a result, manual staffing
projections are made by the District engineers. Additionally, a majority of resident
engineers surveyed said that they do not use CEM MS staffing guidelines in preparing
project staffing plans and primary work schedules. Instead, some indicated that they
staff based on their own experience and judgment, as well as available personnel.
Recent CEM MS field reviews also show that project staffing plans are not submitted
for all projects and that staffing plans are rarely updated.
These f i g u r e s represent only CEMMS costs incurred by central o f f i c e s t a f f and by ADOT
f o r consultants used t o implement, review and modify the system. Although d i s t r i c t
personnel are also involved i n the operation o f CEMMS, t h e i r CEMMS r e l a t e d expenses
a r e n o t included. D i s t r i c t and area operations t e c h n i c i a n s , as well as resident
engineers, spend p a r t of t h e i r time on CEMMS r e l a t e d a c t i v i t i e s .
(') According t o the CEMMS manual, a long- range plan i s developed once a year and covers
a two- year period. The short- range plan i s updated monthly and p r o j e c t s s t a f f i n g
needs f o r t h e next 12 months.
Budget module also not utilized - Although the CEMMS budget module allows for
cost planning, it is not completely understood or fully used by District personnel. The
CEMMS budget module is designed to allow managers to plan costs and monitor
project costs. A budget is developed to plan for costs of construction engineering,
and is based on the work to be done.
Interviews with district management, however, revealed that they were not aware of,
or did not use, the CEMMS budget module. Furthermore, discussions with district
personnel revealed that CE cost estimates, which appear on project staffing plans, do
not always reflect actual expected costs, but instead may result from numbers
backed into the plans to show expressed or implied CE cost limitations. ' we
found that a perceived CE cost limit of 10 percent, for example, caused district
personnel to calculate 10 percent of the project cost and develop a staffing plan
based on that calculation. However, according to ADOT staff, actual staffing needs
may vary significantly from the " 10 percent plan."
CEMMS scheduling function misused and ineffective - ADOT has underutilized and
misused the CEM MS scheduling function. CEM MS provides for short- term ( weekly)
scheduling to assess short- term labor needs, to plan personnel assignments for
efficient use of available staff, and to communicate assignments to employees. A
good plan or schedule has both primary and secondary or alternate work schedules.
However, according to the CEMMS manual, when schedules are not written and
engineers rely on oral communications, crews are not usually deployed effectively.
A survey of construction orgs indicated that more than 25 percent of the orgs do not
prepare primary work schedules weekly. More than 95 percent of the orgs indicated
that they do not post primary work schedules. Primary work schedules are used to
ensure that needed staff are assigned to construction projects.
(' ) A d d i t i o n a l l y , once a construction p r o j e c t i s underway, p r o j e c t charges are sometimes
miscoded i n order to keep CE costs w i t h i n expressed or implied l i m i t a t i o n s ,
p a r t i c u l a r l y on small d o l l a r jobs where ADOT's costs, i n r e l a t i o n t o t o t a l costs, are
higher. D i s t r i c t engineers and central o f f i c e s t a f f have indicated t h a t mi scoding of
p r o j e c t charges i s occurring, due f o r the most p a r t to " expressed and/ or informal
direction" to stay w i t h i n the 15 percent CE cost l i m i t a t i o n on each p r o j e c t .
In addition, some resident engineers do not always accurately reflect their labor
needs on the primary schedules. For example, an ADOT report pointed out that many
schedules are submitted showing 90 percent staff scheduled for project- related work,
yet labor cost distribution reports later show as l i t t l e as 55 percent of the actual
labor costs charged to project related work.
Furthermore, our survey of construction orgs revealed that more than half of the orgs
do not have formal secondary work schedules. More than 77 percent of the orgs
indicated that secondary work assignments were not posted for employees.
Secondary assignments cover productive work of lower priority when the primary
assignments are interrupted by, for example, winter shutdowns, rain or changes in a
contractor's schedule.
Monitoring module underutilized and ineffective - The CEM MS monitoring module -
0
is also underutilized and ineffective. C E M MS currently generates 16 planning and
monitoring reports. The goal in monitoring is to identify significant problems; for
example, to allow resident engineers to identify projects in which man hours or CE
costs excessively overrun initial projections.
However, a survey conducted by our Office found that nearly half of the construction
orgs do not regularly receive CEMMS reports. One area engineer indicated that he
had not received any reports in nearly a year. ADOT district engineers reported that
some districts have decided to stop distribution of CEMMS reports to resident
engineers, which eliminates feedback on labor usage and CE costs. Interviews with
district personnel also revealed that l i t t l e to no formal action is taken once problem
projects are identified.
The System Is Not
Fully Supported
ADOT needs to more fully support CEM MS. Deficiencies and potential improvements
of CEMMS have been pointed out to ADOT management since System
implementation, but l i t t l e action has been taken.
Problems identified - Deficiencies of and potential improvements to CEMMS have
been pointed out to ADOT management repeatedly since 1982, through consultant
reports and internal memorandums. Many issues, such as lack of management
direction, inadequate district staff commitment, system analysis and updating, and
training have been cited on several occasions.
e December 1982 - CEMMS Final Report presented to the ADOT State Engineer by
Roy Jorgensen Associates, Inc., Engineering and Management Consultants.
Jorgensen defined action needed during implementation to bring CEM MS to full
effectiveness. The report also pointed out:
- Need for more involvement of the district engineers.
- Lack of intended district staff commitment to CEMMS.
e November 1983 - CEMMS Progress Report memorandum from the Construction
Engineer to the State Engineer, Chief Deputy State Engineer and Deputy State
Engineer, Highways Operations Group. The memo summarized CE M MS problems
and progress- to- date, and made several recommendations for management
consideration. Points discussed included:
- Need for more involvement of the district engineers.
- Lack of intended district staff commitment to CEMMS.
- Need for clarification of management's desires for central office CEM MS
involvement.
- Need for more central office staff.
- A recommendation for a follow- up review of CEMMS by consultant.
- Need for training.
- Need for more interaction with ADOT1s Information Systems Group ( ISG).
- Problems with project scheduling.
e September 1984 - CEMMS Status Report memorandum from the Construction
Engineer to the State Engineer. The memorandum:
- Expressed concern over the effective use of CEM MS as a management tool
by headquarters and district personnel.
- Recommended a follow- up review of CEM MS by a consultant.
0 January 1985 - CEM MS Review Report prepared by Roy Jorgensen Associates,
Inc., Engineering and Management Consultants. Reviewed CEM MS system
design, data processing and implementation; and made 12 recommendations @
based on their findings. Recommendations included:
- Clarification of CEM MS responsibilities.
- Emphasis on more involvement of central office in management analyses
and setting Statewide manpower levels.
- Central office CEMMS field reviews.
- More analysis at the district level.
- Updates of planning values (') and system software.
0 November 1985 - Suggestions for Improving the ADOT CEMMS COPES
Application submitted by Robert J. Demster, Jr., of Project Management
Systems Consulting, to the CEM MS supervisor. The memo discussed seven major
areas the consultant identified as CEM MS needs. A mong them were:
- Need for training.
- Need to study and update certain CEM MS elements.
- Need for more CEM MS central office staff.
- Need for analysis of field staff information requirements.
0 January 1986 - CEMMS Progress Report memorandum requirements from the
CEMMS Supervisor to the Manager of Field Review Services. The memo cited
deficiencies in several areas that have made CEM MS inefficient, inaccurate, and
of questionable u t i l i t y for project monitoring and labor forecasting. Deficiencies
were noted in:
- Training.
- Lack of Central Office CEM MS field reviews.
- Analysis and updating of planning values and other system components.
- The number of Central Office CEMMS staff available to perform the
required functions.
0 March 1987 - District Recommendations for CEMMS Program presented by
ADOT district engineers to the Deputy State Engineer, Operations Section. The
memo listed the concerns and recommendations district engineers had of the
CEMMS program in five major areas: reliability of reports, coding errors,
schedule reliability, labor projections, and management support. Specific
problems were listed, including:
- Lack of overall direction of the CEM MS program.
- Inadequate central office CEM MS involvement.
- Lack of adequate central office CE M MS staff.
- Need for more interaction with and support from ISG.
- Problems associated with project scheduling.
- Lack of training.
( I ) Planning values are used for guidance i n determining construction engineering labor
requirements. The planning values are expressed i n terms of labor hours per planning
u n i t . For example, a planning value of 96 means that it requires 96 labor- hours to
perform a p a r t i c u l a r a c t i v i t y .
a March 1987 through May 1987 - CEMMS memos from the Assistant State
Engineer, Construction Section, to the Deputy State Engineer, Highway
Operations Group. Three memos were written that discussed current CEMMS
deficiencies, and listed items that had been accomplished, were scheduled or
needed to be addressed to update, improve and utilize the system as originally
intended. Included were:
- Plans for training and field reviews.
- Need for strong commitment to the CEMMS program at all levels.
- Need for greater staff commitment by operations technicians.
- Possible need for more CEMMS Central Office staff.
- Need for system updating.
- Need for more interaction with and support from ISG.
- Need for evaluation of System software.
L i t t l e action taken - Although ADOT management is aware of CEMMS deficiencies
and potential improvements, l i t t l e action has been taken. CE M MS updating has been
recommended since January 1985, but no system updating has been accomplished
since the start of the CEM MS program. ADOT only recently began follow- up field
reviews. Pending requests for additional central office staff are being deferred until
proper evaluation of CEMMS is made and the system is fully operational. Problems
with inadequate district staff commitment still exist. Furthermore, strong
commitment and need for management determination of staff duties and
responsibilities and overall program direction still need to be addressed.
ADOT Needs To Take Action
To Make CEMMS Viable
ADOT needs to address System deficiencies if CEMMS is to become a viable
management tool. Important system elements need to be maintained, and
management issues should be addressed i f CEM MS is to become effective.
System elements - lmportant system elements need to be maintained. Consultant
reports and internal ADOT memos point out that CEMMS should be periodically
updated. The CEMMS manual contains guidelines for reviewing and updating the
system annually to ensure that it is kept current and in tune with ADOT's work
methods. However, critical system elements have not been updated. For example,
planning values were initially determined in 1982 from estimates of ADOT senior
resident engineers. ADOT expected to update the planning values based on actual
information once the program was underway, but this has still not been done. Five
year- old estimates continue to be used. Failure to update planning values is one
reason cited by district engineers for CEM MS ineffectiveness in projecting manpower.
Management issues - Management issues should also be addressed i f CEM MS is to
become an effective manpower management tool. Central office's role in Statewide
construction staff management should be restored. Additionally, management should
address issues dealing with inadequate staffing and inadequate support from A D 0 T's
Information Systems Group.
Central office's role in Statewide labor management should be reasserted. Roy
Jorgensen Associates said that an important part of CEMMS implementation is
involvement of the central office staff in overall Statewide staffing. When CEM MS
was implemented it was used by central office to balance classifications and staffing
for the State. District engineers, however, strongly objected to having central office -
make these decisions for them. In January 1985, Jorgensen found that central office
had discontinued that practice. Control of CEMMS was turned over to the districts.
Highway Operations management indicates that central office's role is to assist the
districts in managing their labor resources. As a result, there is l i t t l e Statewide
management of CE staff. Moreover, central office feels that the districts are
responsible for managing their CE labor and, therefore, see l i t t l e need to monitor or
take action when problems arise.
Additionally, management should address inadequate central office staffing for
CEMMS. Although central office has added staff in the last year, there are still some
questions as to their capacity to perform major CEMMS functions. In 1986, the
CEM MS supervisor requested six additional positions to adequately staff three major
functions: CEMMS training and technical support, data processing, and audits and
analysis. Management also needs to reduce the amount of non- CEMMS time to which
i t assigns CEMMS staff. District engineers cited reassignment of CEMMS central
office staff by senior management to tasks unrelated to CEMMS as one reason
CEMMS staff have been unable to perform some of their functions. They reported
that 20 to 50 percent of CEMMS central office staff time has been spent on duties
not envisioned at the inception of the program.
Finally, ADOT upper management needs to address the lack of adequate lnformation
Systems Group ( ISG) involvement and support for CEMMS. When CEMMS was being
designed, ADOTts lnformation Systems Group declined to assist in the effort because
the software CEMMS uses is a vendor purchased system. The system's continued
function and maintenance was to be worked out between the vendor and CEMMS
central office staff. However, ADOT has had to hire a consultant on several
occasions to modify system software. According to CEM MS staff, lack of ISG
support needs to be addressed by ADOT management. ADOT's ISG is part of the
Administrative Services Division, while CEMMS is located in the Highways Division.
Highways Division management has indicated that they will seek firm corn mitments
from ISG for support of the CEMMS System.
RECOMMENDATIONS
1. ADOT should strongly support the CEMMS program, requiring full utilization by
both central office and district management.
2. ADOT should provide adequate resources to properly carry out the CEM MS
program at the central office, district and area levels.
3. The ADOT director should require that sufficient support be given CEMMS by
the A DOT lnformation Systems Group.
4. A D 0 T should:
e Address current CEMMS deficiencies and direct that adequate analysis and
modification of the System be accomplished.
e Clearly define the CEM MS program goals and objectives.
e Specify the responsibilities of each organizational level, from senior
management to field staff.
a Reassert central o f f i c e management's role in setting Statewide labor levels.
a Make provisions for more management analysis at both the central office
and d i s t r i c t levels.
OTHER PERTINENT INFORMATION
During the audit other pertinent information was developed regarding the cost of
consultant managed construction projects.
Consultant Managed Construction Projects
Consultant managed ( C M) CE costs have varied when compared to ADOT CE costs
for similar- sized projects. On large projects, CM costs have been lower. However,
on small and medium- sized projects CM costs have been higher. ADOT, however,
could realize certain benefits that may offset any extra costs i f the Department
appropriately employes engineering consultants.
A D 0 T contracts with engineering consulting firms to perform construction
engineering functions on a limited number of projects. Reasons for entering into
these contracts are twofold: 1) for ADOT to meet peak workload demands, and 2) to
assist ADOT when the Department does not have the technical expertise to manage
certain specialized projects. The department has entered into at least 30 contracts
of this nature during the last three years.
Since the number of consultant managed projects is small, CE cost comparisons with
ADOT managed projects must be considered tentative. "' ( See Table 9) Among
the two largest categories of construction projects the CE cost percentages of
consultant managed projects are lower than ADOT managed projects. Within the four
remaining project size categories for which cost comparisons can be made, however,
CE cost percentages are higher for consultant managed projects. A t times, these
differences are considerable.
( I ) Cost comparison i s a1 so d i f f i c u l t because consultants may charge costs which ADOT
cannot. On Federal p a r t i c i p a t i o n highway projects, a consultant can charge a
pro- rated portion of the f i r m ' s overhead costs to construction engineering. ADOT i s
not permitted to do t h i s on projects managed in- house a1 though the Department incurs
s i m i l a r overhead costs.
ADOT could realize certain benefits that may offset any extra costs on medium and a
smaller projects i f the Department appropriately utilizes engineering consultants.
Instead of hiring additional construction staff during workload peaks, consultants
could be employed as needed. In such instances, ADOT only has to contract for
additional construction staff when needed and would not be required to release or a
transfer employees when the workload returns to normal.
TABLE 9
CONSTRUCTION ENGINEERING COST PERCENTAGE COMPARISONS
OF COMPLETED CONSULTANT AND ADOT MANAGED CONSTRUCTION PROJECTS
FISCAL YEARS 1983- 84 THROUGH 1986- 87
CE Percentage CE Percentage
Project Size Consultants Number I n- House Number
$ 10 Mi I l i o n + 7.04% 2 8.70% 2
$ 5- 9.99 Mi I l ion 6.51 1 9.38 13
$ 2- 4.99 M i l l i o n 15.99 5 10.11 2 4
$ 1- 1.99 M i l l i o n 17.15 4 11.77 5 5
$ 500,000- 999,999 17.80 4 11.37 54
$ 1 00,000- 499,999 28.72 5 16 - 36 132
Source: Auditor General analysis of ADOT construction costs
AREAS FOR FURTHER AUDIT WORK
During the course of our audit we identified potential issues that we were unable to
pursue because they were beyond the scope of our audit or we lacked sufficient time.
a Could the claims documentation process be improved?
Several ADOT employees expressed concern that contractor claims for highway
construction may be excessive. A December 1986 report indicated that the total
potential liability to ADOT for outstanding claims was approximately $ 24
million. Of the $ 24 million, ADOT paid out $ 10 million ( 42 percent of the total
liability). In addition, a January 1987 report revealed that ADOT has
approximately $ 14 million in active contractor claims.
Further audit work is needed to determine why the amount of claims is so high
and what can be done to address the problem.
a Should A DO T1s design process be improved?
Numerous Highways Division staff and one contractor have indicated that
construction design plans are inadequate or of poor quality, and lead to many
problems in the field. One ADOT official attributed poor plans to ADOT's
current accelerated project schedule. Others indicated that there is inadequate
coordination of design plans on projects that have multiple design consultants.
Additionally, ADOT staff stated that poor design plans can result or have
resulted in an excessive number of contract change orders, force accounts and
claims.
Further audit work is needed to verify staff complaints about the design process
and to determine what can be done to minimize design problems in the future.
a Could A DOT streamline its organizational structure and eliminate unnecessary
levels of supervision?
ADOT may be able to streamline its organizational structure. In fiscal year
1982- 83 the Department reduced the number of engineering districts from seven
to four in an effort to improve administrative efficiency. Although some
positions were eliminated and districts were consolidated, another layer of
administration was added in the outlying districts with the creation of nine area
offices. A review of the current staffing levels and reporting structure in the
districts reveals that spans of control and numbers of staff reporting to area
engineers vary among districts. For example, one area engineer supervises 11
orgs, while another supervises only three. In addition, personnel under the area
engineers range from approximately 36 to 151 staff. Furthermore, there are at
least five levels of supervision in some districts between the district engineer a
and field crews, yet numbers of org staff are relatively small in some cases. -
Further audit work is needed to determine if the current organizational structure
and reporting hierarchy is necessary and justified.
0 Could ADOT's project scheduling process be improved?
Discussions with ADOT staff, as well as ADOT documentation and a consultant
report, indicate that the Department's project scheduling process needs
improvement. Field construction staff have stated that planning labor needs
becomes more difficult when project schedules are inaccurate. A 1985
consultant report indicated that ' I. . . manpower plans will continue to be
inaccurate and show excessive manpower needs until the construction letting
schedule becomes more accurate." In addition, two ADOT internal memos have
revealed problems with ADOT1s project scheduling.
Problems with scheduling, among other things, have led the districts to override
C E M MS staffing projections with a manually generated manpower report.
District engineers have said that they have l i t t l e faith that project scheduling
will become substantially more accurate than it currently is. Further audit work
is necessary to determine to what extent the project scheduling process is
inaccurate and whether it can be improved.
I)
ARIZONA DEPARTMENT OF TRANSPORTATION
206 South Seventeenth Avenue Phoenix, Arizona 85007
I) October 29, 1987
EVAN MECHAM CHARLES L MILLER
Governor D~ rector
Mr. Douglas R. Norton
Auditor General
2700 N. Central Ave., Suite 700
Phoenix, Arizona 85004
Dear Mr. Norton:
The following comments constitute our formal response to the
most recently revised draft of the performance audit of the
~ rizona Department of Transportation's construction Management
function. The structure and format of our response is divided
into an introduction and general comments, a specific response
to each finding, a detailed comment of your supporting analysis,
and a specific response to each recommendation.
INTRODUCTION AND GENERAL COMMENTS
The report consists of two findings. The first alleges that ADOT
has not adequately controlled construction staffing levels, and
the overstaffing cost the state $ 7.4 million over the last four
fiscal years. We totally reject this findinq. Further, the in-ference
that $ 7.4 million has been wasted or resulted in exces-sive
costs to the State of Arizona is not supported by facts.
The second finding states that ADOT1s construction Engineering
Manpower Management System is not being utilized effectively and
needs greater management support. The Construction Engineering
Manpower Management System ( CEMMS) is a computerized manpower
planning and monitoring tool. We concur that it can be used more
effectively, and we anticipate the application of additional re-sources,
plus management support, through organizational change.
In order to provide background information that will be germane
to the discussion which follows, it is necessary to note that
significant change has occurred in the programming level of
construction projects over the last four fiscal years. The size
of the ADOT construction program has increased significantly.
@ d
- 1-
). HIGHWAYS AERONAUTICS . MOTOR VEHICLE PUBLIC TRANSIT AOM~ N~ STRATIVSEE RVICES TRANSPORTAT~~ PNL ANNING
These changes resulted from a series of funding mechanisms at
both the federal and state levels beginning in 1983- 84 as a
result of the Federal 1982 Surface Transportation Assistance
Act. During the same period, significant increases were achieved
with an incremental increase in state gasoline taxes, resulting
from legislative action in 1982.
Beginning in January 1986 as a result of the previous year's
legislation, additional funding was provided for use statewide,
and as a result of the half- cent sales tax increase in Maricopa
County major construction funding was initiated. These changes
resulted in a 155 wercent increase in construction contracts
advertised and awarded over a five- year period of time. Over
that same period the number of employees assigned to field
construction contracts varied by only 11.4 percent. The number
of full- time construction FTEs per $ 1 million worth of constant
dollar construction was reduced from 5.9 in 1983 to 2.65 in
1987. This indicator of efficiency and productivity per construc-tion
employee is shown graphically in Figure 1.
FTE'S PER MILLION $
FISCAL YEAR
LEGEND
This increased efficiency resulted from a number of innovative
techniques developed by the department during that period of
time. They significantly affect the analysis and will be dis-cussed
in detail further in our response.
SPECIFIC RESPONSE TO FINDING I
The Arizona Department of Transportation does not concur with
Finding I. Further, we totally reject the conclusion drawn as a
result of the finding. Even if we were to agree with the sub-stance
of your methods, manner and conclusion ( and we do not),
we could not conclude, nor do we agree, that the resources con-stituted
by such alleged overstaffing were not productive for
the State of Arizona. The inference that $ 7.4 million of costs
incurred by the State as a result of the alleged overstaffing is
not supported by fact. This conclusion infers a waste of man-power,
and that has not been the case.
DETAILED COMMENT
The analysis included in the audit report employed three separ-ate
methods tabulated in Tables 2, 3 and 5 labeled " The Con-struction
Engineering Cost Methodw, " Construction Staff
Productivityff, and m miscellaneous Time Charged Method1' respec-tively.
The results purport to demonstrate overstaffing from
1983 through 1987, ranging from 78 FTEs to 387. You selected
the Construction Engineering Cost Method as being the most
appropriate.
You noted that utilization of these three separate methods of
analysis help to validate your results. It should be pointed out
that the three methods utilized in your report yield results
that vary from 37 percent to 613 percent. We do not believe such
comparison indicates a high probability of verification.
However, we fully recognize the difficulty of finding a specific
method that can be utilized in arriving at an appropriate level
of construction manpower. Each method can indeed be used if
appropriately applied. Each method is analyzed below with appro-priate
input variables and comment where the analysis is flawed.
construction Ensineerins Cost Analysis
In employing the Engineering Cost Analysis you have utilized
basic data and information that is utilized in CEMMS. For
example, the construction engineering cost goals, which are
established for each project and used as a budgeting guide
by the Department for engineering manpower, were extracted
from each project and applied to specific year's contractor
payments. While the system itself is not designed to yield
precise numbers of employees to achieve those CE cost goals,
it is true one can utilize such an approach, together with
cost per hour data, and back into a calculation of equiva-lent
average number of FTEs per year.
However, any system or formula devised to audit the efficien-cy
of CEMMS in the above analytical setting must be based
upon a common foundation with CEMMS. Unfortunately, your
analysis is not.
Normal manpower planning assumes sufficient employees to per-form
specific tasks within a given time period. The accepted
practice is to initiate manpower planning on the base of
2,080 hours per year per full- time employee ( 40 hours/ week x
52 weeks/ year = 2,080). The standard 2,080 hours/ year must
be adjusted for annual and sick leave and miscellaneous
time.
Historical data shared with your office indicates that with-in
ADOT an average of 291 hours/ year per employee is charged
against sick and annual leave. Further data, also shared
with your office, indicates that an additional 208 hours- per
year per employee are charged to non- construction and other
valid, productive assignments not directly related to con-a
struction. Theref ore, the total number of hours/ year per
employee available to construction activities in CEMMS is
1,581 ( 2,080 - 291 - 208 = 1,581). This figure must be used
for " Hours per Yearw if an accurate analysis is to be per-formed.
This differs significantly with the value of 1,789
hours per year available as per your calculation in Table 2.
Thus, the results obtained utilizing the more appropriate
1,581 hours per year as noted above are set forth in Figure
2.
f
a
Figure 2
I Construction staff as determined by an analysis of Construction
Engineering ( CE) costs for FY 1983- 84 through 314 of FY 1986- 87
' 2 ) Fxc'cies c r o j e c : ~ a6nin; s: ered by private col: uitinc firns. ( Auoitor Generai cetermined)
( 5': OvzrCr;: gas: baser. on cummuiarive apolicsr! on of ACGT Ci ccsr coals. ( Aucitor Genera! ileterrined)
( T: Ifi$ i- c! ete~? intx averege cost per man hour for the iiscal year. ( 8asec 07 C E V S stzpdarc of ZO8O hrs/ year)
I, ,, T : b ~ c ~ 7ai. li 7~ b.?+ r 0 ; t ? ~ . : : r ~ ~ ~ r i o ~ n ~ c i n en~ ourrisn gwo rkeb per year by one emoioyee. ( Base of 2080 hrslyear) .:-', , 9305 72: : T:.;: CE ~ t t ~ nee'r s: t z cversee cc: su;? ai~ r - 2;:;:: 9r: je', t:.
, . $ ; 2 e : . - . -
, , -': :: r: t: i , c - i c r o v e r s e ~ c o x u ; t d ~ t mrn3ce: ~ " r j e z ; ~ .( Aul'tc- Geqcre; d e c e v i ~ e c ' ,
Utilizing the same data and methodology as set forth in your
Table 2 with this one change, we find that the Department
was not overstaffed, with the exception of FY 1985- 86. In
this instance, projects were not advertised and awarded as
originally scheduled, because approximately $ 45 million
worth of projects were delayed. This delay resulted from
application of the Grahamm- Rudman- Hollings bill and budget
deficit procedure at the federal level. We readily admit we
did not adjust our work force for this short period of time,
because of the complexities of training and acquiring
construction inspection personnel.
As shown in Figure 2, ADOT has used CEMMS properly to plan
and control our construction staff over the past four years.
In fiscal years other than 1985- 86 ADOT has been slightly
understaffed in construction, resulting in a total opportun-ity
cost savings of $ 1,869,379 as shown in ~ igure 3.
r
1 Figure 3 I
I COST OF COXSTRUCTION STAFF FISCAL YEARS 1 9 8 3 - 8 4 THROUGH 1986- 87 I
FISCAL STAFF AVBRAGE STAFF WAGE COST
YEAR DIFFERENCE WAGE COST DIFFERENCE
2. Productivity Method of Calculation
In addition to the above- noted consideration regarding the
average yearly staff hours available, the analysis labeled
" Staff Productivityw in Table 3 of your report utilizes
another factor which warrants comment. The productivity
standard factor ($ 163.17) is noted to be the highest level
of productivity achieved by ADOT to date. That productivity
was achieved in the fiscal year just completed and validates
the productivity and efficiency increases shown in our Fig-ure
1. However, this analysis requires some further comment.
During the past four fiscal years significant innovative
techniques have been implemented by the Department, which
affect the construction engineering manpower efficiency. The
Department has developed and initiated specifications requir-ing
" end resultff acceptance of Portland cement concrete and
asphaltic concrete, totally changing the manpower required
for the Department to monitor and control the production of
these two materials. As a result of those specification
changes, the construction industry itself has grown increas-ingly
more proficient in the control of its own operation.
Both have resulted in a lesser amount of manpower required
in behalf of the Department.
During the same period, the Department initiated and imple-mented
ffcontractor stakingff of construction projects. This
has reduced significantly the construction manpower require-ments.
The significant point here is that the productivity
achieved in the past year is not a level that would have
been applicable to previous years, because of the changes
made in the method in which these activities were carried
out. Thus, the application of the productivity factor of
$ 163.17 to years prior to the current year yields results
which are inaccurate.
3. Miscellaneous Time Analysis
Inasmuch as the miscellaneous time factor, when utilized and
reported properly, is an indicator of only those hours of
productive time spent on non- direct project- chargeable
activities, we agree it is a tool for utilization by the
local level manager as well as Central Office staff to judge
activities, but it is not an indicator of efficiency.
To summarize, we believe our analysis set forth in Figure 2 is
the most realistic approach that can be achieved when utilizing
these methods. From the information available to the Department,
we do not believe a finding of overstaffing is valid with the
exception of the one year indicated. Retention of trained staff
until additional work became available was an entirely appropri-ate
approach to the management of the construction engineering
manpower.
RECOMMENDATIONS
Following is the Arizona Department of Transportation's response
relative to the recommendations for Finding I.
1. ADOT should consider increasing its use of alternative staff-ing
methods as needed to handle peak construction periods
statewide and in seasonal districts before adding additional
permanent staff.
Response :
- Use of consulting ~ ngineers
The Arizona Department of Transportation will continue to
utilize consulting engineers as staff extensions where appro-priate.
We deem this activity to be appropriate where peak
workloads exist and, more importantly, where unique construc-tion
activity requires expertise which we do not maintain
within our staff. We do not, however, anticipate expanding
greatly the utilization of consulting engineers, because
this approach is not the most cost- effective or proficient
in terms of professional service rendered.
- Contracting Out of Functions
ADOT currently provides for " contractor stakingw and has for
the past two years. Certain laboratory functions and inspec-tion
are also contracted out. We will continue these
practices.
- Use of Temporary Staff
We have in $ he past and will continue to utilize temporary
staff .
- District Transfers
We have in the past and will continue to utilize staff
transfers where appropriate.
- Share Resources With ~ aintenance
While we have used construction personnel to supplement
maintenance, particularly in the winter season, there is
very little opportunity to reverse that trend and utilize
maintenance work forces for construction. This inability
results from the training required for construction inspec-tion
technicians as well as the staffing of maintenance
crews which require a full complement based upon the annual
maintenance plan.
2. The Legislature should consider setting and monitoring ADOT
construction staff levels.
Response :
The Department does not oppose working with the Legislature
on this matter. It must be realized, however, that a signifi-cant
degree of flexibility would be necessary in order to
meet the fluctuations in construction funding which occur on
extremely short notice-- particulary in the federal program.
For example, the Department received approximately $ 299 mil-lion
in additional discretionary federal funds for Inter-state
construction over the past few years. These funds must
be placed in construction status within 90 days of receipt.
Accordingly, the normal budget cycle is such that initial
budget requests are prepared and submitted 23 months before
the end of the fiscal year in which they are to be utilized.
3 . The Legislature should consider acquiring a followup perform-ance
audit in two years of ADOT1s construction management
function to review its progress in managing staff size.
Response :
The Department welcomes a followup review of this activity
at any time.
RESPONSE TO FINDING I1
Finding I1 deals with the Construction Engineering Manpower
Management System ( CEMMS) utilized by ADOT. We agree this
system could be utilized more effectively. The CEMMS system was
initiated by the Arizona Department of Transportation in 1982.
Like many other systems developed in the last decade to assist
in managing transportation and highway department operations,
the basic goal has been to take normal management techniques
together with appropriate criteria and automate them through the
use of software and computers. The validity of such systems,
however, and thus their usefulness, depends upon their simpli-city.
In other words, to be utilized effectively and willingly by man-agement
at all levels such systems must provide a useful tool
without overburdening them with either input or output data.
With this in mind, it is the intent of ADOT to review critically
the overall function of the CEMMS system with the objective of
strengthening it. Further, it will be the objective of manage-ment
to ensure proper utilization of all applicable line man-agers.
RECOMMENDATIONS
ADOT should strongly support the CEMMS program, requiring
full utilization by both central office and district manage-ment.
-
Response :
ADOT has supported CEMMS in the past and will continue to
increase that support in the future through organizational
changes, which are being considered.
ADOT should provide adequate resources to properly carry out
the CEMMS program at the central office, district and area
levels.
Response :
The approach we intend to take in improving CEMMS utiliza-tion
should require minimal increased resources. However,
those required will be made a budget issue for FY 1988- 89.
3. The ADOT director should require that sufficient support be
given CEMMS by the ADOT Information Systems Group.
Response :
Where data processing systems modifications and programming
are required by CEMMS, ADOT management will utilize ISG
personnel or outside expertise as deemed to be most cost
effective.
4. ADOT should: * Address current CEMMS deficiencies and direct that
adequate analysis and modifiction of the System be
accomplished. * Clearly define the CEMMS program goals and objectives. * Specify the responsibilities of each organizational level,
from senior management to field staff. * Reassert central office management's role in setting
statewide labor levels. * Make provisions for more management analysis at both the
central office and district levels.
Response :
All five objectives outlined in the above recommendation
will be accomplished.
Sincerely, I_- -- ,.'" . I
L '. Charles L. Miller
Director