Skip Repetitive Navigation Links
California State Auditor Report Number : 2015-120

California Department of Transportation
Its Maintenance Division’s Allocations and Spending for Field Maintenance Do Not Match Key Indicators of Need

Audit Results

After Beginning to Develop a Logical Approach for Addressing Field Maintenance Needs, the Maintenance Division Abandoned It

Through its prior budget model (model) and other planning documents, the California Department of Transportation (Caltrans) had laid out a logical approach for assessing and addressing field maintenance needs.5 This included categorizing all state highways into specific zones based on climate and traffic volume. It also included calculating maintenance performance scores (service scores) for maintenance activities in each zone and setting goals for improving them. In concept this information should logically inform where maintenance needs are greatest and therefore how funding should be allocated and prioritized. However, the maintenance division abandoned this approach. In practice the maintenance division identifies field maintenance work based primarily on physical observations made by maintenance personnel who drive along sections of district roads and the division’s reactions to public complaints. Further, instead of using the more localized zone approach for evaluating service scores, it uses a much higher‑level evaluation by calculating an overall service score for certain maintenance activity categories each district is responsible for and averaging the district scores to derive statewide scores. In addition, the maintenance division only sets performance goals for improving three of the many categories of maintenance activities it reports on annually. It also has based annual funding allocations to the 12 districts on a simple average of historical spending rather than using service scores or zone‑level information, despite reporting to the Legislature that it was using a more sophisticated method. Thus, it is not surprising that we found actual spending has not improved service scores and has been inconsistent among roadways in similar zones.

The Maintenance Division Does Not Use Key Indicators and Performance Information to Strategically Plan Field Maintenance Activities

The Federal Highway Administration, other state departments of transportation, and additional entities have established that traffic volume, climate, and environmental factors are key factors that influence highway maintenance needs. Further, Caltrans’ maintenance manual states that the need for maintenance activities, such as routine litter removal and road sweeping frequencies, is very sensitive to traffic volumes. In 2009 the maintenance division developed a budget model for its field maintenance program that accounted for some of these key factors. Specifically, that model would have allowed the maintenance division to identify the resources needed to maintain highways with similar climate and traffic volume at a given maintenance performance level. However, despite paying a consultant roughly $250,000 to develop the model, the maintenance division never implemented it. According to the deputy director for maintenance and operations (deputy director)—who was the chair of the steering committee responsible for overseeing the consultant’s efforts in developing and implementing the model—the model suggested unreasonable allocations to the districts, such as reducing district 7’s (Los Angeles) staff by roughly 100 positions, so the maintenance division abandoned it.

In concept, the model would have shifted the maintenance division’s assessment of maintenance need from a district‑level analysis to a more granular and informative route‑level analysis. After dividing all state highways, or routes, into segments by county, the model grouped these segments into one of five climate regions and six traffic volume categories, as shown in the text box. Generally, roads with higher traffic volume have greater field maintenance needs. For example, the model categorized the segment of Interstate 80 in Sacramento County as having inland valley climate and the highest traffic volume, level 5 (Inland‑5). The model referred to each combination of climate region and traffic volume category, such as Inland‑5, as zones. The model established service scores at the zone level and, using historical trends of expenditures and labor hours, estimated the resources needed to maintain those scores. Finally, using this zone‑level information and data about the inventory of assets subject to maintenance activities, the model would have established baseline funding needs and allocations to maintain similar highways within the state to a specified service score.6

Highway Zone Climate and Traffic Categories

Climate Region

Traffic Volume (average daily traffic)

Source: California Department of Transportation’s budget model User Manual and Technical Guide.

However, the maintenance division did not implement the zone‑level evaluation of service scores it had envisioned in its abandoned budget model. While that approach would have allowed it to monitor each district’s maintenance performance by zone, its current methodology of sampling roads for evaluation does not provide this level of information. Specifically, the current methodology gathers information only for each district as a whole and does not accurately identify differences in maintenance needs that exist within each district. As mentioned in the Introduction, the maintenance division calculates service scores for many maintenance activities but only sets service score goals for three of them: litter and debris, guardrails, and striping. In addition, these goals apply only to overall statewide performance, not to individual districts.

In Figure 4 we present the statewide service goals and service scores for the categories of maintenance activities for which the maintenance division has established service score goals. We also include the service scores for each of the three districts we reviewed: district 4 (Oakland), district 6 (Fresno), and district 7 (Los Angeles). The maintenance division calculates the statewide service scores by averaging the respective scores of each district. From fiscal year 2010–11 through 2014–15, the state average service score was below the goal for striping and guardrails, but was above the goal for litter and debris. However, the service scores for the three districts we reviewed were generally below the state average. Moreover, although service scores for these districts have been volatile, scores for districts 4 and 7 have generally been well under the state average, as Figure 4 shows.

However, the maintenance division does not weight districts’ service scores to account for differences in traffic, terrain, or climate, even though the maintenance performance statewide report for fiscal year 2014–15 stated that readers should consider these characteristics when comparing scores for each district. For example, districts 4 (Oakland) and 7 (Los Angeles), which handle almost 43 percent of the State’s traffic volume, have the same impact on the overall state service score as do districts 1 (Eureka) and 9 (Bishop), which together handle about 1 percent of the State’s traffic. Evaluating maintenance performance without including traffic volume allows the several small districts with minimal traffic to outweigh the few large districts with the majority of the State’s traffic, skewing the statewide score higher. When we weighted the score for fiscal year 2014–15 to reflect the traffic volume in each district, we found that although the statewide score for guardrails did not change, the scores for striping and litter and debris worsened by between five and nine points.


Figure 4
Service Scores for Categories With Service Goals for Districts 4 (Oakland), 6 (Fresno), and 7 (Los Angeles)
Fiscal Years 2010–11 Through 2014–15

A chart showing districts 4 (Oakland), 6 (Fresno), and 7 (Los Angeles) service scores for the maintenance categories of litter and debris, guardrail, and striping from fiscal years 2010-11 through 2014-15.

Sources: California Department of Transportation’s Maintenance Level of Service Statewide Reports Executive Summaries for fiscal years 2010–11 through 2014–15.

Instead of holding individual districts accountable for meeting service score goals for field maintenance activities, the maintenance division established spending goals for specific activities that the districts monitor as indicators of performance. When we inquired as to how these spending goals were derived, the division was unable to support how the spending goals were originally established—it only asserted that they are adjusted based on past expenditures and service scores. According to the assistant maintenance division chief for the office of administration and budgets (assistant division chief), these spending goals provide direction to the districts on improving service scores. Although there is value in monitoring spending, this approach alone does not account for the actual activities that need to be completed or the outcomes from that spending that would result in improved service scores. In fact, our review found instances where the districts’ service scores decreased even though the districts sometimes exceeded their spending targets. For example, despite spending more than its target for litter and debris since fiscal year 2012–13, district 7’s (Los Angeles) service scores for this category declined.

Despite the fact that the service scores for the districts we reviewed have generally not met the statewide goals, the maintenance division could not demonstrate that it is using this information to strategically plan its work to address maintenance needs and improve service scores. While the maintenance division’s service score evaluation guide states that managers can use service scores in conjunction with their own knowledge and other evaluation data to develop plans and set priorities, we found that districts do not use the service scores to identify and prioritize their field maintenance work. Deputy district directors of maintenance (district maintenance deputies) we spoke with stated that they do not use service scores to inform how they plan or prioritize the field maintenance work they perform. Because districts are not using information regarding field maintenance needs to plan and prioritize, their field maintenance work appears to be performed on an impromptu basis and is not consistent with need.

In addition, the three districts we reviewed generally create only short‑term plans for activities they intend to complete during the following one or two weeks; such plans do not adequately establish priorities or facilitate the monitoring of progress toward improving service scores. For example, district 6 (Fresno) provided us with emails that supervisors sent to superintendents that briefly described activities they intended to complete over the next week. In the 2005 version of its maintenance manual, the maintenance division included requirements for districts to formally plan and monitor field maintenance work using the maintenance management system and requirements for region managers and district maintenance deputies to review annual workload plans. The manual also required districts to take into account service scores at the route level when developing plans.

These plans would have enabled management to establish and document priorities and develop a baseline by which to measure progress. For example, the maintenance manual specifies that districts are to establish routine litter removal and sweeping frequencies for each highway segment based on the rate at which litter, debris, and sediment accumulate. However, according to the division chief, the maintenance division subsequently removed the field planning requirements from the maintenance manual because districts determined that using the planning and scheduling module in the maintenance management system was very time‑consuming. Despite the significant benefits of formally planning field maintenance work, the maintenance division did not develop alternate planning methods that it believed were more feasible. Without adequate plans for completing field maintenance work, district staff are not being held accountable for how well they address field maintenance needs, as evidenced by the low service scores and the inconsistent spending discussed in a later section.

The division chief also asserted that maintenance crews cannot always adhere to work plans because they are regularly redirected from completing planned work to address emergency work, such as clearing accident scenes and addressing other urgent highway issues. However, we do not agree with that rationale for not planning work. If crews cannot strictly follow their plan because of the responsive nature of maintenance work, these deviations from the plan will shed light on what work activity is not getting accomplished and what changes to resource allocations should be made. While we agree that the maintenance division must respond to emergencies and may need to interrupt or reschedule routine work, those interruptions make it especially important to plan and prioritize the maintenance division’s other work to ensure that it is adequately addressing field maintenance needs.

The Maintenance Division Does Not Allocate Field Maintenance Funding to Its Districts Based on Key Indicators of Need

The maintenance division does not use key indicators of field maintenance need, such as traffic volume and service scores, to allocate funding to the districts. Thus, some districts may not be receiving adequate funding to meet their field maintenance needs, which may delay the work and result in more costly future repairs. We found that the proportion of maintenance funds the maintenance division allocated to each district remained roughly the same for fiscal years 2010–11 through 2014–15, as Table 5 shows.


Table 5
California Department of Transportation Division of Maintenance Allocations of Maintenance Program Funding to the Districts and Headquarters
Fiscal Years 2010–11 Through 2014–15
FISCAL YEAR 2010-11 FISCAL YEAR 2010-11 FISCAL YEAR 2010-11 FISCAL YEAR 2010-11 FISCAL YEAR 2010-11
District 1 (Eureka) $43,029,662 3% $32,867,925 3% $47,611,739 4% $45,393,086 4% $42,266,069 4%
District 2 (Redding) 66,373,821 5 61,677,143 5 61,817,647 5 64,908,119 6 70,069,479 6
District 3 (Marysville) 115,878,610 9 100,368,446 9 92,961,662 8 109,275,871 9 116,896,285 10
District 4 (Oakland)* 154,851,477 12 150,110,958 13 172,561,926 15 151,492,677 13 160,405,815 13
District 5 (San Luis Obispo) 52,509,804 4 53,308,742 5 36,521,774 3 47,253,414 4 57,951,401 5
District 6 (Fresno)* 80,203,232 6 81,585,820 7 81,710,979 7 81,469,870 7 93,469,564 8
District 7 (Los Angeles)* 164,458,341 13 173,443,490 15 145,577,384 13 151,959,677 13 167,490,827 14
District 8 (San Bernardino) 107,998,998 9 108,706,246 10 93,875,323 8 85,494,829 7 106,996,691 9
District 9 (Bishop) 30,340,056 2 27,484,149 2 30,572,337 3 31,253,822 3 33,158,160 3
District 10 (Stockton) 64,175,575 5 64,369,844 6 57,533,387 5 61,511,945 5 69,905,348 6
District 11 (San Diego) 77,077,839 6 66,541,799 6 57,718,317 5 65,765,906 6 75,703,342 6
District 12 (Irvine) 47,130,705 4 47,533,874 4 43,821,146 4 45,119,848 4 42,691,895 4
Headquarters 244,176,880 20 172,298,564 15 208,315,953 18 210,704,765 18 161,870,124 14
Totals $1,248,205,000 100% $1,140,297,000 100% $1,130,599,574 100% $1,151,603,829 100% $1,198,875,000 100%

Sources: California Department of Transportation’s financial system.
* We selected district 4 (Oakland), district 6 (Fresno), and district 7 (Los Angeles) for detailed review as part of our audit.
The maintenance division’s allocation of maintenance funding to headquarters was primarily to the following four divisions: division of maintenance, division of engineering services, division of procurement and contracts (specifically, warehouse), and division of audits and investigations.
Percentages do not add to exactly 100 percent for all fiscal years due to rounding.

At most, the percentage of allocations to each of the districts changed by zero to two percentage points year to year. The allocations for headquarters had slightly wider swings; its allocations ranged from a high of 20 percent in fiscal year 2010–11 to a low of 14 percent in fiscal year 2014–15. Although these allocations generally aligned with the number of lane miles in each district, they were not commensurate with other indicators of maintenance need. One key indicator of each district’s maintenance need is traffic volume, which was a key element of the abandoned budget model and, as previously mentioned, a key indicator used by several other governmental entities.

Figure 5 shows that although some districts have a significantly higher proportion of traffic volume, the maintenance division does not allocate a commensurate proportion of funding to those districts. For example, as previously mentioned, together districts 4 (Oakland) and 7 (Los Angeles) handle 43 percent of the State’s traffic volume, but these two districts together received only 27 percent of maintenance program’s funding in fiscal year 2014–15. On the other hand, multiple districts with significantly lower traffic volume received disproportionately large allocations. For instance, districts 1 (Eureka), 2 (Redding), and 9 (Bishop) together received 13 percent of total program funding in fiscal year 2014–15, despite only handling 3 percent of the state’s traffic volume. Although traffic volume is not the only indicator of maintenance need, the large differences between districts’ proportions of traffic volume and allocations suggest that the maintenance division should revise its allocation methodology.

The maintenance division also does not allocate funding based on districts’ service scores. Consequently, some districts may not have the resources to improve those scores. As we describe in the Introduction, the maintenance division annually determines service scores to measure districts’ field maintenance performance. The maintenance division’s performance evaluation guide states that one important objective of evaluating performance is to assess statewide maintenance needs. However, although service scores at some districts were generally low over the last five fiscal years, the maintenance division did not revise those districts’ overall allocations to address those low service scores. For example, from fiscal year 2010–11 through 2014–15, service scores for litter and debris, striping, and guardrails at districts 4 (Oakland) and 7 (Los Angeles) have, for the most part, been significantly below both the state goal and the state average.

Because the maintenance division does not allocate its funding commensurate with key indicators of maintenance needs, some districts may not be able to fully address these needs, which can delay maintenance work and result in more costly repairs in the future. This effect is exacerbated because headquarters is potentially underfunding those districts where a significant portion of the State’s highway traffic is occurring and that have low service scores.


Figure 5
Percentages of Maintenance Program Funding Districts Received in Fiscal Year 2014–15 Compared to Districts’ Percentages of 2013 Daily Vehicle Miles of Travel on State Highways

 A chart showing the maintenance division’s 2014-15 allocations to the 12 Caltrans districts compared to the 12 districts’ 2013 daily vehicle miles of travel on state highways.

Sources: California State Auditor’s analysis of allocation data for fiscal year 2014–15 from California Department of Transportation’s (Caltrans) financial system and daily vehicle miles of travel on state highways data from Caltrans’ 2013 California Public Road Data report.
* We selected district 7 (Los Angeles), district 4 (Oakland), and district 6 (Fresno) for detailed review as part of our audit.
Caltrans headquarters received 14 percent of maintenance program funding in fiscal year 2014–15.
District 9 (Bishop) handles 0.4 percent of the State’s daily vehicle miles of travel on state highways.

The Maintenance Division Has Mischaracterized the Sophistication of the Method It Uses to Allocate Field Maintenance Funding to the Districts

Excerpt from the California Department
of Transportation’s 2015
Five‑Year Maintenance Plan

The Maintenance Program Budget Model was developed to enhance budget management capabilities on an annual basis. The performance‑based model uses a combination of historical expenditures, Level of Service performance measures, and inventory data to project future resource needs with performance‑level expectations for the entire State Highway System. While the budget model does include all resources, the model is used primarily to determine a performance‑based budget for field maintenance activities.

Historically, the budget model grouped Caltrans’ 12 districts into sets of comparable units using geographic, population, and traffic volume characteristics. Current versions of the model have shifted the focus from district‑level analysis to route‑level analysis by grouping all routes based on traffic volumes and geographic locations. Through route‑level analysis, the budget model provides detailed comparative analysis for determining relationships between performance and resource needs for each highway inventory unit which creates standardized allocation and efficiency rates for each route.

The budget model is used:

Source: California Department of Transportation’s 2015 Five‑Year Maintenance Plan (maintenance plan).
Note: The 2011 and 2013 maintenance plans contained the same descriptions of the budget model, except for minor word changes.

Although it did not implement the model that it developed for its field maintenance activities, the maintenance division has been reporting to the Legislature and other decision makers that it was using the model’s sophisticated methodology for allocating field maintenance funds. As we previously noted, the maintenance division has continued to allocate funding in roughly the same proportions from fiscal year 2010–11 to 2014–15. For example, district 7 (Los Angeles) has received between 13 and 15 percent of the total allocation in each of the last five fiscal years. Although the maintenance division’s process for allocating funding to the districts is not well documented, the maintenance division asserted that it has allocated field maintenance funding to the districts using the number of positions in each district and averaging the districts’ historical expenditures. The maintenance division annually informs districts of their allocations and instructs districts not to exceed them. We verified with the three districts we reviewed that they do not estimate and propose allocations to the maintenance division. The deputy director and the division chief stated that they do not know when the maintenance division originally started using this allocation methodology or how it originally determined the number of staff positions and the funding allocations for field activities for each district.

However, Caltrans consistently reported that it was using the model “to determine a performance‑based budget for field maintenance activities” in the 2011, 2013, and 2015 updates to its five‑year maintenance plans (maintenance plan). As shown in the text box, in these plans Caltrans also provided additional details and reported that “through route‑level analysis, the budget model provided detailed comparative analysis for determining relationships between performance and resource needs for each highway inventory unit which created standardized allocation and efficiency rates for each route.” However, as previously mentioned, according to the deputy director, the maintenance division has never used the model to calculate route‑level allocations to the districts for the entire field maintenance budget. Consequently, the Legislature and other decision makers may have believed that the maintenance division was using a more sophisticated approach to allocate funding to the districts than it actually has, and thus they may have been less willing to challenge or question the allocations. The division chief stated that his staff will revise the language in the subsequent maintenance plans to more accurately depict how the maintenance division currently allocates funding for field activities to the districts.

Although the division chief agreed that allocating funding for field activities using traffic volume and service scores would result in allocations commensurate with districts’ needs, he stated it would not improve large urban districts’ service scores because these districts would not be able to spend their increased allocations. According to the deputy director and the division chief, large urban districts—such as districts 4 (Oakland) and 7 (Los Angeles)—cannot fill many of their field staff positions because of staff turnover resulting from the high cost of living, terms in union agreements that affect how the program advertises field positions, and the department’s lengthy hiring process. However, if indicators of maintenance need show that funding allocations should be adjusted, we believe the maintenance division needs to work on resolving any challenges that it encounters. In addition, our analysis showed that the vacancy rates in the three districts that we reviewed remained consistently low. Specifically, the average vacancy rate for authorized positions was 3 percent or less from fiscal year 2012–13 through 2014–15. In fiscal year 2014–15, districts 4 (Oakland), 6 (Fresno), and 7 (Los Angeles) on average had 912, 325, and 876 authorized positions, respectively. Additionally, Caltrans asserted that it allocates blanket positions to the districts, and when we considered that information, average vacancy rates at the reviewed districts did not increase significantly and ranged from 4 percent to 10 percent in fiscal year 2014–15.7 According to the assistant division chief, Caltrans is reviewing its existing hiring processes and policies to expedite the hiring process for field maintenance staff.

The Maintenance Division Does Not Spend Its Field Maintenance Funds Consistent With Some Indicators of Need

Given the shortcomings with the maintenance division’s approach to planning and allocating funds we discussed in previous sections, we were not surprised to find that actual spending for field maintenance work was not consistent with key indicators of need. Our review of Caltrans districts 4 (Oakland), 6 (Fresno), and 7 (Los Angeles) found that each district spent different amounts for field maintenance on highways with similar maintenance needs based on key indicators Caltrans identified—traffic volume and climate. For example, each of the three districts spent different average amounts per mile on certain highways that had similar field maintenance needs. In addition, we found that each district spent significantly less on some highways in its district than it spent on others with similar maintenance needs. While other indicators of maintenance need may also affect spending, the inconsistent spending we identified suggests that the maintenance division should assess whether it is effectively addressing maintenance needs. By potentially underspending, it increases the personal safety and environmental risks posed by unmet maintenance needs, such as excess litter or graffiti, guardrails in need of repair, unfilled potholes, and untended landscaping. Further, unperformed maintenance activities can result in more significant future repair costs.

When we used the abandoned model’s zone methodology to compare average spending per mile during our audit period, we found districts sometimes spent significantly different amounts for field maintenance on highways with similar maintenance needs. For example, each of the three districts we reviewed has highways in an inland valley climate with traffic volume levels 1 through 4. Table 6 shows that district 7 spent roughly twice as much per mile on highways classified as Inland‑3 as did districts 4 and 6: $154,000 compared to $71,000 and $62,000, respectively. In addition, district 7 spent more than $1 million per mile to maintain its four‑mile‑long section of Inland‑2 highway, while it spent only $334,000 per mile to maintain its Inland‑5 highways. According to the district maintenance deputy for district 7 (Los Angeles), the high spending along this section of Inland-2 highway was driven by work including landscaping, highway sweeping, and removal of debris and graffiti, which required costly lane closures. Table A in the Appendix shows the average spending per mile for each zone in the three Caltrans districts we reviewed. Because highways with similar characteristics should give an indication of similar maintenance and resource needs regardless of which district they are in, these spending variances among the districts may mean that the districts are not spending sufficient resources to meet maintenance needs, not receiving sufficient resources, or they are not spending their resources efficiently.

For the three reviewed districts, we plotted the maintenance division’s field maintenance expenditure data from fiscal years 2010–11 through 2014–15 on highway zone maps to see how each district’s field maintenance spending corresponded to maintenance need. Specifically, we first calculated the average dollars that each district spent per mile in each of the zones. Then we compared sections of highways that are in similar zones and identified those sections where the district spent below 50 percent of the average. We considered this to be low spending. We also identified high spending for each zone as more than 150 percent of the average spent per mile in the respective zone. As part of our analysis, we reviewed SHOPP projects planned since 2000 that rehabilitated the pavement and might have reduced the amount of spending for field maintenance needed on the highways we identified. We found that SHOPP projects had been performed on some roads with low spending as well as on some roads with high spending; thus, it appears that the inconsistent spending is not entirely explained by those projects.

Table 6
Average Spending per Mile in Inland Valley Climate Zones Fiscal Years 2010–11 Through 2014–15
Climate / Traffic Zone Average Dollars Spent Per Mile Per Zone
Climate Traffic Volume* District 4 (Oakland) District 6 (Fresno) District 7 (Los Angeles)

Inland Valley

1 (0–2,500)
$22,000
$47,000
$79,000

Inland Valley

2 (2,501–10,000)
41,000
32,000
1,007,000

Inland Valley

3 (10,001–25,000)
71,000
62,000
154,000

Inland Valley

4 (25,001–100,000)
90,000
115,000
107,000

Inland Valley

5 (100,001+)
159,000
NA
334,000

Sources: California Department of Transportation’s (Caltrans) integrated maintenance management system and Caltrans’ climate region and traffic volume data.

Note: Average dollars spent per mile are weighted for mileage and rounded to the nearest thousand.

NA = District 6 (Fresno) does not have any highways in the inland valley climate with a traffic volume 5.

* Traffic volume indicates average daily traffic.

Figure 6 identifies in blue the sections of highways on which district 4 spent less for field maintenance than it spent on other highways in the same district with similar maintenance needs. For example, the blue text box in Figure 6 shows a section of an approximately 41‑mile stretch of Route 82 in district 4 that is in the North Coast‑4. On average, district 4 spent $104,000 per mile to maintain North Coast‑4 roads. However, it spent only an average of $28,000 per mile for field maintenance on this particular section of Route 82, despite this section having maintenance needs similar to the district’s other North Coast‑4 roads. This is a difference of 73 percent—or almost $76,000 less per mile in spending—compared to other similar roads in district 4. Similarly, Figure 6 shows in red the sections of highways in district 4 with high spending. In particular, the red text box in Figure 6 shows district 4 spent $66,000 more per mile than the average of $56,000 for its Low Mountain‑3 highways, or $122,000 per mile on an 11‑mile section of Route 116 in Sonoma County, which is more than twice the average. Both district 6 and district 7 also spent field maintenance funds inconsistently on highways with similar maintenance needs. Figures 7 and 8 show sections of highways in districts 6 and 7, respectively, on which these districts spent more than 150 percent of the average or more than 50 percent below the average.

Figure 6
Field Maintenance Per‑Mile Spending, District 4 (Oakland) Highways

A map showing areas of high and low field maintenance spending on segments of highways in district 4 (Oakland).

Source: California State Auditor geographic information system consultant’s analysis of cost data from California Department of Transportation’s (Caltrans) integrated maintenance management system for fiscal years 2010–11 through 2014–15 and Caltrans’ climate region and traffic volume data.
* Traffic volume 3 indicates average daily traffic of 10,001–25,000.
Traffic volume 4 indicates average daily traffic of 25,001–100,000.

Figure 7
Field Maintenance Per‑Mile Spending, District 6 (Fresno) Highways

A map showing areas of high and low field maintenance spending on segments of highways in district 6 (Fresno).

Source: California State Auditor geographic information system consultant’s analysis of cost data from California Department of Transportation’s (Caltrans) integrated maintenance management system for fiscal years 2010–11 through 2014–15 and Caltrans’ climate region and traffic volume data.
* Traffic volume 3 indicates average daily traffic of 10,001 – 25,000.
Traffic volume 4 indicates average daily traffic of 25,001 – 100,000.

Figure 8
Field Maintenance Per‑Mile Spending, District 7 (Los Angeles) Highways

A map showing areas of high and low field maintenance spending on segments of highways in district 7 (Los Angeles).

Source: California State Auditor geographic information system consultant’s analysis of cost data from California Department of Transportation’s (Caltrans) integrated maintenance management system for fiscal years 2010–11 through 2014–15 and Caltrans’ climate region and traffic volume data.
* Traffic volume 3 indicates average daily traffic of 10,001–25,000.
Traffic volume 5 indicates average daily traffic of 100,001+.

We also compared the three districts’ spending with their median income and their race and ethnicity demographics and found there was no clear correlation. In the Appendix, we describe our analysis and present the socioeconomic demographic information as additional context for the districts’ spending. Specifically, in the Appendix, Figures A, B, and C show the median income and Figures D, E, and F show the demographics for the predominant racial and ethnic makeup for the three districts we reviewed.

As part of our review, we brought the field maintenance spending discrepancies to the attention of the districts, and we asked for their perspective on why field maintenance spending varied among highways with similar needs in their respective district. The district maintenance deputies indicated that several factors could be affecting the spending, such as previous SHOPP projects that reduced the need for field maintenance, the age of the road, whether the road is in a rural or urban area, the number and cost of lane closures required to perform the field maintenance work, the number of service requests received for a particular road, and the amount of landscaping on a particular road. However, as we described previously, the maintenance division does not consider these factors in its planning or allocating for field maintenance work. Although we are not suggesting that Caltrans spent money on projects that were not worthwhile or where there was no indicator of need, we are concerned that there are many highways in each district on which average spending per mile over a five‑year period was inconsistent with the maintenance needs indicated by the nature of the climate and traffic volumes in those zones.

The Maintenance Division Cannot Demonstrate That It Promptly Performs Field Maintenance Work

The three districts we reviewed do not effectively manage the service requests they receive from the public to ensure that they appropriately prioritize and address them in a timely manner. Specifically, data indicate that more than 30,000 service requests received by the three districts in fiscal years 2010–11 through 2014–15 remained unresolved after more than 90 days. Although these districts asserted that many of those service requests have been completed, neither we nor the districts’ management could easily verify this assertion. Unperformed maintenance work can create safety risks and can also result in increased future repair costs.

The public can identify needed maintenance work by using an electronic form on Caltrans’ website that allows the public to notify the maintenance program of a specific maintenance issue—for example, graffiti, litter, or a pothole—and its location. The maintenance division tracks these service requests in its statewide maintenance service request system (service request system). After receiving a service request from the public, a district assesses the request to decide how it should respond. Specifically, the district determines whether the reported issue occurred on the state highway system and if the district needs to address it. After conducting this assessment, the district responds to the individual who submitted the service request, stating whether the district commits to perform the work or stating the reasons why it is not going to perform the work. Although the maintenance division has not established a time frame for completing service requests, we analyzed the service requests that the maintenance division had committed to addressing by grouping them in 30‑day increments to determine how long they had been outstanding. Data from the service request system suggests that two of the three districts we reviewed frequently let service requests remain outstanding for more than 90 days, as shown in Table 7. For example, from fiscal year 2010–11 through 2014–15, more than 16,000 service requests submitted to district 7 remained outstanding for more than 90 days. Similarly, during the same time period, the system data showed that more than 15,000 service requests submitted to district 4 had not been addressed after more than 90 days.


Table 7
Status of Maintenance Service Requests Received by California Department of Transportation Districts 4 (Oakland), 6 (Fresno), and 7 (Los Angeles)
Fiscal Years 2010–11 Through 2014–15
District 4 (Oakland) District 6 (Fresno) District 7 (Los Angeles)
  Number of service requests* Percent of district’s service requests Number of service requests* Percent of district’s service requests Number of service requests* Percent of district’s service requests
Completed service requests:
Completed in 30 or fewer days
14,893
41%
1,103
77%
1,400
7%
Completed in 31 to 60 days
3,116
9
57
4
121
1
Completed in 61 to 90 days
823
2
51
4
62
0.3
Completed in more than 90 days
1,295
4
148
10
184
1
Outstanding service requests:
Outstanding for 30 or fewer days as of June 30, 2015
413
1
2
0.1
327
2
Outstanding for 31 to 60 days as of June 30, 2015
377
1
2
0.1
227
1
Outstanding for 61 to 90 days as of June 30, 2015
376
1
5
0.3
388
2
Outstanding for more than 90 days as of June 30, 2015
15,288
42
61
4.3
16,783
86
Totals 36,581 100% 1,429 100% 19,492 100%

Source: California State Auditor’s analysis of data from California Department of Transportation’s maintenance service request system.
* In this analysis we included only those service requests for which the districts committed to perform the work.
Percentages do not add to exactly 100 percent for all fiscal years due to rounding.

The districts’ current processes for tracking service requests make it difficult and time‑consuming to determine whether maintenance staff have completed service requests and if data in the service request system are accurate. Districts use the service request system to track service requests from the public while district maintenance staff use the maintenance management system to record that they completed the work associated with these service requests. However, the two systems do not automatically exchange information between one another, and the districts have to manually update the service request system to show which service requests were completed. Otherwise, the data on the status of service requests could be inaccurate. District maintenance deputies could not verify whether the large number of service requests that were outstanding for more than 90 days had been completed, and they indicated that doing so would require a significant effort. Nevertheless, we believe the maintenance division should have a process for following up on those service requests that it does not promptly address.

Furthermore, the districts we reviewed do not capture in the service request system those service requests they receive via methods other than Caltrans’ website, making it even more difficult to monitor the progress of addressing those additional service requests. The district maintenance deputies at the three districts we reviewed confirmed that their districts do not have documented policies that describe how they should handle service requests received by phone, mail, or email. Although some districts have tools to record such service requests, none of the districts we reviewed record such requests in one central repository. For example, some regions in district 4 (Oakland) use paper logs to record phone calls. In another example, according to the district maintenance deputy for district 6 (Fresno), the district does not have any process to record and monitor service requests it receives by phone, mail, or email. As a result of the districts’ poor processes for recording service requests they receive via methods other than Caltrans’ website, neither we nor the three districts’ staff could ensure that such service requests were addressed in a timely way or addressed at all.

Similarly, the maintenance program does not have processes to ensure that maintenance staff complete work orders in a timely manner. As we described in the Introduction, the maintenance manual requires maintenance staff to create work orders in the maintenance management system for all field maintenance work they perform. The system has fields to record the dates when a work order is initiated and when it is completed, and these dates can be used to determine how long it took to complete the work. However, according to the office chief of management systems and studies within the maintenance division, supervisors do not consistently create work orders as soon as they identify field maintenance needs; rather, they often create work orders after their crews start the work. Thus, the true length of time it takes from the date the work is identified to the date it is completed is not captured, and the maintenance division cannot accurately monitor whether it is addressing field maintenance needs in a timely manner.

Weak Controls Over Field Maintenance Work Orders Create Opportunities for Fraud, Waste, and Abuse

Caltrans’ weak controls over field maintenance do not adequately ensure that work order costs are reasonable and allowable and that the resources used were necessary and appropriate. Our review of field maintenance work orders found that the integrated maintenance management system is paperless and no supporting documentation is maintained for work order costs, such as labor, equipment, and materials used to complete field maintenance work. We reviewed the compensating internal controls the maintenance division has developed to ensure that costs are reasonable and allowable. The maintenance manual describes controls including superintendent reviews of work orders, which are intended to monitor work order costs. Internal controls are essential to ensure that the maintenance division achieves its objectives and mitigates the risk of fraud, waste, and abuse. Maintenance supervisors record on each work order all costs incurred to complete the work. For example, a supervisor records in the work order how many hours each staff member works and the costs of all materials and vehicles used to complete maintenance activities. Although the maintenance manual requires superintendents to review work orders for material usage and accurate time reporting, there is no evidence that they do so. In particular, according to the office chief of management systems and studies, the current maintenance management system does not allow reviewers to document that they have reviewed a work order. Without such evidence, the maintenance division cannot ensure that the labor, equipment, and materials used are reasonable and appropriate.

We also reviewed Caltrans’ internal division of audits and investigations’ (internal audits) reviews or audits of these costs and found that the internal auditors had identified weaknesses. For example, a 2015 Caltrans internal review of district 8 (San Bernardino) found weaknesses in controls over materials such as the chemicals used for controlling vegetation. Specifically, internal audits’ review noted variances in 20 of 40 inventory counts between the count and the inventory records for those items. The review indicated that when physical inventory levels are not reconciled to balances recorded in the maintenance management system, theft or misuse of the chemicals can go undetected. Internal audits made recommendations to address these issues but has not yet followed up to determine whether the recommendations have been implemented. Additionally, an internal audit released in 2013 found that some maintenance staff entered erroneous time sheets into the system, including time sheets that exceeded available leave balances and time sheets showing total numbers of hours worked that did not match the employees’ work schedules. The audit also found that staff failed to promptly correct errors and that superintendents did not review and approve time sheets until the end of the month, which, according to the internal auditors, could result in errors and misuse of leave. Caltrans’ internal auditors made recommendations for these findings that the maintenance division has reported addressing.

The Maintenance Division Appropriately Manages Highway Maintenance Projects, but Some Backlogged Maintenance Work Is Increasing

In contrast to the field maintenance activities, the maintenance division currently uses a needs‑based approach to allocate funding for the majority of highway maintenance activities and to identify highway maintenance projects. The maintenance division allocates funding for highway maintenance proportionally to the districts based on the condition of the pavement and bridges in each district. For example, it allocates funding for bridge projects based on each district’s portion of outstanding bridge maintenance work that regular bridge inspections identify. Similarly, it determines districts’ allocations for pavement projects using data about the amount and condition of the pavement in each district. In contrast, according to the office chief for storm water and environmental compliance, the maintenance division generally allocates highway maintenance funds for drainage and culverts equally among the districts because the budget is relatively small compared to the identified needs of the districts. Currently, the maintenance division generally identifies projects to repair pavement by conducting field reviews and considering data from the pavement condition survey, which uses customized vehicles fitted with sensing equipment traveling on the roadway at or near highway speeds to collect pavement condition data. Based on these pavement surveys, each lane mile is rated as good, fair, or poor. Further, the maintenance division stated that it is in the process of implementing a new technology called PaveM that it indicates will optimize the project selection process by targeting future repairs that provide the best value for the least amount of money. According to the 2015 State of the Pavement report, PaveM will make suggestions based on pavement condition, pavement type, climate, and project history. However, the division has not set a final implementation date for this new technology.

Our review of highway maintenance projects for pavement and bridges found that generally the maintenance division ensured that contractors completed projects within budget and on time. While the maintenance division lacks strong controls for field maintenance, controls for highway maintenance projects to contain costs generally appear to be adequate. For highway maintenance projects that contractors perform, Caltrans’ resident engineers review contract payment requests before authorizing contractor payments, which entails comparing completed work with the contract. We found that contractors generally completed the 15 projects we reviewed at an actual cost that was reasonably close to their bid amounts and within the number of days estimated in the bid.

However, the maintenance division’s goals to repair pavement statewide fail to address its growing backlog of needed pavement maintenance. Specifically, pavement that is in fair condition with minor surface distress, such as minor cracking and potholes, but that still requires corrective maintenance has increased. According to Caltrans’ State of the Pavement reports, the backlog of lane miles in need of corrective maintenance has increased from 11,053 in 2011 to 15,272 in 2015 due to the increase in construction costs and a greater percentage of higher‑cost maintenance strategies. In Caltrans’ 2015 maintenance plan, the maintenance division established a goal to repair 2,100 lane miles a year based on the funding it received, although Caltrans reported that it did not meet this goal.

The maintenance division is also not sufficiently addressing backlogged inspections and repairs of drainage culverts even though culvert damage or failure can seriously damage roadways, create the need for extensive repairs, and threaten the mobility and safety of the traveling public. In the 2015 maintenance plans, Caltrans set goals to inspect 12,000 culverts and repair 140 culverts annually. Caltrans reported that it inspected 13,168 culverts in fiscal year 2013–14, which exceeded its goal. Although Caltrans has generally met its culvert repair goals during our audit period, the goal to repair 140 culverts annually does not sufficiently address the significant increase in backlogged maintenance needs. The number of culverts Caltrans reported as needing repair has consistently grown since fiscal year 2010–11, from an estimated amount of 13,185 to 27,346 as of June 2015. The backlog is increasing in part because Caltrans’ culvert inspection program has identified and assessed the condition of 107,000 of the estimated 205,000 culverts in the State. As this process continues, Caltrans is identifying more culverts in need of repair. Additionally, Caltrans reports that a large percentage of culvert inspections and repairs it identified through previous maintenance plans have included “easier” access and repairs, and it anticipates that the remaining culvert inspections and repairs will be more difficult to address and will require additional time and planning to complete.

According to the division chief, Caltrans has taken an approach of developing SHOPP projects that focus on repairing pavement and culverts that are in the worst condition first. In fact, Caltrans reported in its 2015 maintenance plan that if it were to receive additional revenue, it would prioritize SHOPP pavement rehabilitation ahead of preventative maintenance. Caltrans also reported that it would recommend funding both maintenance and SHOPP to manage and maintain backlogged culverts. Given the need to balance resources between SHOPP and maintenance, we believe Caltrans should continue to monitor and report on its backlogged maintenance work to ensure that the backlogged work does not deteriorate to such a degree that it results in increased future SHOPP costs.

In contrast to its lack of progress on pavement and culverts, in its 2015 Five‑Year Maintenance Plan Caltrans reports that it is reducing its backlogged highway maintenance projects for bridges. During our audit period Caltrans set a goal to reduce the number of bridges with backlogged maintenance needs to 8 to 10 percent of the total inventory of bridges. The bridge inventory reportedly varied between 12,900 and 13,100 during our audit period. Caltrans defines bridge maintenance needs as backlogged if two years pass without completion of the maintenance after a bridge inspector recommends maintenance work. During our audit period, Caltrans reported that bridges with backlogged maintenance needs decreased from 2,575 in the beginning of fiscal year 2010–11 to 1,771 in the beginning of fiscal year 2013–14. Caltrans reports it is consistently reducing the backlog and estimates that the backlog will continue to decrease at the current level of funding. In 2015 Caltrans set a goal to reduce the number of backlogged bridges to 1,090 over the next five years.

Recommendations

To better align the maintenance division’s allocations with districts’ maintenance needs, the Legislature should include language in the Budget Act that requires the maintenance division to develop and implement a budget model for field maintenance by June 30, 2017, that takes into account key indicators of maintenance need, such as traffic volume, climate, service scores, and any other factors the maintenance division deems necessary to ensure that the model adequately considers field maintenance need. Once the model is developed, Caltrans should use it to inform appropriate allocations to the districts.

Caltrans should revise the language in its future five‑year maintenance plans to accurately describe the method it uses to allocate field maintenance funding to its districts.

To ensure that it performs field maintenance work consistently on highways with similar needs, the maintenance division should do the following:

Caltrans should require its staff to verify and update the status of all outstanding service requests. Additionally, Caltrans should require supervisors to monitor completion of service requests by reviewing the data from the service request system monthly to identify service requests not completed after a period of time that Caltrans deems appropriate, such as 30 days. For all service requests outstanding after this period, Caltrans should require its supervisors to determine the status of the service request by reviewing the related work order that records what work Caltrans completed and ensure the work is appropriately prioritized. Also, Caltrans should require its staff to record all service requests it receives via methods other than Caltrans’ website, such as by phone, mail, or email, in its service request system to ensure it captures all service requests in one central repository.

To detect and prevent fraud, waste, and abuse and to ensure costs are appropriate, the maintenance division should strengthen its controls over reviewing and approving work order costs by requiring its supervisors and superintendents to document their review and approval of work orders in the maintenance management system. For example, supervisors or superintendents could include a note in the comment field of the work order indicating their review and approval. The maintenance division could also establish a reasonable dollar threshold for those work orders that would require documented review and approval.

To ensure that field maintenance work orders are completed in a timely manner, the maintenance division should require supervisors to initiate work orders in the integrated maintenance management system at the time that they identify field maintenance work that needs to be performed and record the date that work was started and the date the work was completed. Superintendents should periodically review work orders to ensure that identified work is completed in a timely manner.

We conducted this audit under the authority vested in the California State Auditor by Section 8543 et seq. of the California Government Code and according to generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives specified in the Scope and Methodology section of the report. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Respectfully submitted,

ELAINE M. HOWLE, CPA
State Auditor

Date:
March 17, 2016

Staff:
Nicholas Kolitsos, CPA, Audit Principal
Jordan Wright, CFE
Martin Lee, CPA
Caitlin Palmer
Natalja Zvereva

IT Audits:
Michelle J. Baur, CISA, Audit Principal
Lindsay H. Harris, MBA, CISA
Ryan P. Coe, MBA, CISA
Shauna M. Pellman, MPPA, CIA

Legal Counsel:
Heather Kendrick, Sr. Staff Counsel

Geographic Information System Consultant:
ENPLAN

For questions regarding the contents of this report, please contact
Margarita Fernández, Chief of Public Affairs, at 916.445.0255.



Back to top


Footnotes

5 Five students we reviewed moved out of the SELPAs we selected for this audit before the 2012–13 school year (the second year after AB 114 took effect). For those students, we reviewed only changes to mental health services that occurred in the school year immediately following when AB 114 took effect.Go back to text

6 The U.S. Department of Education required the California Department of Education to submit its annual performance report in February 2015 using data from the 2013–14 school year.Go back to text

7 For Murrieta Valley, we reviewed contractors hired by its SELPA office, Riverside, because, according to the special education director at Murrieta Valley, the LEA does not hire its own contractors but instead uses those hired by its SELPA.Go back to text