Indiana 2006 Five Percent Report

This report is in response to the Federal requirement that each state describe at least 5 percent of its locations currently exhibiting the most severe highway safety needs, in accordance with Sections 148(c)(1)(D) and 148(g)(3)(A), of Title 23, United States Code.  Each state’s report is to include potential remedies to the hazardous locations identified; estimated costs of the remedies; and impediments to implementation of the remedies other than costs.  The reports included on this Web site represent a variety of methods utilized and various degrees of road coverage.  Therefore, this report cannot be compared with the other reports included on this Web site.

Protection from Discovery and Admission into Evidence—Under 23 U.S.C. 148(g)(4) information collected or compiled for any purpose directly relating to this report shall not be subject to discovery or admitted into evidence in a Federal or State court proceeding or considered for other purposes in any action for damages arising from any occurrence at a location identified or addressed in the reports.

Additional information, including the specific legislative requirements, can be found in the guidance provided by the Federal Highway Administration,
http://safety.fhwa.dot.gov/safetealu/fiveguidance.htm.

---

HIGHWAY SAFETY IMPROVEMENT PROGRAM
INDIANA “5 PERCENT REPORT”

This report is to meet the requirement for Indiana to submit an annual report describing not less than 5 percent of their highway locations exhibiting the most severe safety needs. The intent of the report is to raise the public awareness of the highway safety needs and the challenges that exist.

Extent of Coverage

The crashes used for this report are for the calendar years of 2003, 2004 and 2005 obtained from the Indiana State Police. The extent of coverage is for Interstate, US and State Roads, those administered by INDOT. The only intersections included are those at intersections of INDOT to INDOT administered roads. The report does not include local roads.

The crash records were assigned first to some 1085 intersections if the crash fell within 250 feet of the intersection. The remaining crashes were assigned to segments of which there were 19,262.

Fatal and injury crashes were combined and assigned. All injury crashes were used since at this time serious injury crashes are difficult to distinguish.

Explanation and an Estimated Schedule of when Full Coverage will be Reported

The present report is based on the approximate 11,700 miles of INDOT administered network and does not include local roads. There was not time to develop a complete road network. An investigation will be made to determine what road networks are available. The state wide transportation model does contain a road network of approximately 30,000 miles which will be considered. It is also a possibility that the state and local road inventory network could be used which would give 100% coverage. At any rate next years report will be expanded considerably.

Approximately 80% of the crashes had sufficient information that they could be matched to an intersection or road segment. Processes being developed will improve on the number of crashes that can be assigned.

Methodology Used to Determine 5% most Hazardous Locations

For each intersection and road segment five measures of safety and evaluation criteria were calculated. These are the fatal-injury Annual Crash Frequency, fatal-injury Annual Crash Density, fatal-injury Crash Rate, fatal-injury Index of Crash Frequency and the fatal-injury Index of Crash Cost.

Appendix B contains a Working Report that explains these measures in detail and the procedures used to produce them.

Because the index of crash costs represents a measure of the impact that crashes have on society it was chosen as the factor to determine Indiana's segments and intersections that exhibit the most severe safety needs. Those intersections and segments that had 4.0 standard deviations more crash cost than could be expected were chosen. The number of fatal and injury crashes for these selected intersections and road segments were then summed. It was found that they represent approximately 7.2 percent of the fatal-injury crashes for 2003 through 2005. Therefore these intersections and roadway segments represent more then the minimum requested.

There were 93 roadway segments selected out of the 19,262 roadway segments analyzed and 11 intersections out of 1,085. This shows that 0.51% of the segments and intersections analyzed accounted for 7.2% of the fatal-injury crashes.

A listing of the intersections and road segments can be found in appendix A.

Assessment of Potential Remedies, their Estimated Costs, and Impediments to their Implementation

Detailed improvement remedies and costs are currently not available for each segment or intersection. Each road segment and intersection was compared to projects either planned or under construction. Under the potential remedy column an “A” indicates that the road segment or intersection needs to be reviewed, an “X” means that a project presently exists that will mitigate the safety need. There were several locations where projects are scheduled but since the project does not necessarily mitigate the safety need an “A” was assigned and the safety need will be reviewed. Included in the table are Project Designation numbers (DES#), improvement type and the estimated cost. Projects that are already planned will be reviewed to see if the proposed project can incorporate mitigation of the safety need.

Approximately 30 of the severe safety need locations have projects planned that will mitigate the safety need. Another 74 locations did not and will be evaluated and programmed for future project if practical.

The above mentioned listing will be shared with the districts for consideration in the next call for projects and as projects develop, the costs and remedies will be included in the annual “5 percent report”.

5% Locations for Indiana

Route	Seg/ Int	Ref Post From	Ref Post To	Description	County	District	Potential Remedy	Projects - Scheduled or Recently Let
								DES #	Estimated Yr of Cn	Project Type	Cost $(millions)
U-012	SEG	0+00	0+06	From Illinois St line to the SW ramp B to I-90	LAKE	LaPorte	A	0600003	2010	HMA	.804
S-045	SEG	38+13	38+28	From the N Jct with SR 37 East 0.15 mi	MONROE	Seymour	A	NONE
S-144	SEG	7+00	7+20	From the Morgan Co Ln to SR 37	JOHNSON	Seymour	X	9611710&0500397	2006	Int Imp	1.172
I-065	SEG	7+36	8+26	From SR 60 to.0.84 mi S of SR 311	CLARK	Seymour	X	9241965	2006	ATR	10.114
U-040	SEG	67+57	68+09	From 0.30 mi W of then to the Marion Co Ln	HENDRICKS	Crawfordsville	A	8562070	2006	ATR	9.818
S-933	SEG	0+00	0+26	From the Elkhart Co Ln W to the Osceola Corp Ln	ST JOSEPH	LaPorte	A	NONE
S-048	SEG	35+87	35+93	From SR 37 W 0.60 Mi	MONROE	Seymour	A	NONE
U-031	SEG	101+31	102+50	From Hilltop Dr W to the Marion Co Ln	JOHNSON	Seymour	A	0101375	2006	Partial 3R	.333
U-030	SEG	6+85	7+11	From 0.03 mi W of Pine Isl St to Scherriville/ Merrillville City Limits	LAKE	LaPorte	A	NONE	2007	HMA
U-006	SEG	27+70	27+84	From 0.32 mi E of SR 46 to "Old Hwy SR 49"	PORTER	LaPorte	A	NONE
S-026	SEG	38+56	38+65	From 0.9 Mi W of Creasy Ln to Creasy Ln in Lafayette	TIPPECANOE	Crawfordsville	A	NONE
S-019	SEG	114+92	114+96	From 0.12 mi S to 0.08 S of Franklin St in Elkhart	ELKHART	Fort Wayne	X	9801130	2009	3R/4R	17.859
S-002	SEG	13+69	13+82	From SR 35 to SR 4 in LaPorte	LA PORTE	LaPorte	A	0300497	2004	HMA	.955
U-031	SEG	160+78	160+87	From .09 mi S of Lincoln Rd to Lincoln Rd in Kokomo	HOWARD	Greenfield	A	0300866&0200094	2006	Partial 3/R	4.359
U-031	INT	162+41		US 31 at US 35	HOWARD	Greenfield	A	NONE
I-065	SEG	0+89	1+19	From 0.43 mi to 0.13 mi S of Stanisfer Ave in Jeffersonville/Clarksville	CLARK	Seymour	A	9966920
U-030	SEG	4+99	5+06	From 0.01 mi W of to 0.06 mi E of Cline Ave in Shererville	LAKE	LaPorte	A	NONE	2007	HMA
U-031	SEG	129+50	129+69	From 0.11 mi N to 0.30 mi N of 146th St in Westfield	HAMILTON	Greenfield	A	NUMEROUS
U-040	SEG	102+67	103+53	From CR 600 E to 0.03 mi E of CR 700 E	HANCOCK	Greenfield	A	0400522
S-111	SEG	35+60	35+34	From 0.34 mi S of SR 60 to SR 60	CLARK	Seymour	X	9902540	2008	3R/4R	22.143
S-037	SEG	171+33	171+44	From 0.01 mi S of 141st St to 0.10 mi N of 141st St	HAMILTON	Greenfield	X	0400361	2015	ATR	59.760
S-252	SEG	0+00	0+15	From SR 37 E 0.15 mi.	MORGAN	Seymour	A	0600272
S-025	SEG	72+92	73+00	From 0.08 mi S of SR 17 to SR 17	CASS	LaPorte	X	0300773	2009	3R/4R	5.000
U-031	SEG	69+50	69+73	From Beam Road to SR 46 (25th St) in Columbus	BARTHOLOMEW	Seymour	X	9700230	2008	ATR	25.667
I-074	SEG	34+05	39+36	From the Bridge over CSX RR (0.13 mi East of US231) to SR 32 NE of Crawfordsville	MONTGOMERY	Crawfordsville	A	0501241	2008	HMA	8.000
S-018	SEG	103+24	103+49	From Western Ave to D St. in Marion	GRANT	Fort Wayne	A	NONE
U-030	SEG	11+04	11+10	From 0.06 mi W of then to Mississippi St in Merrillville	LAKE	LaPorte	A	NONE	2007	HMA
U-024	SEG	68+66	71+96	From 1.36 mi E of to 2.16 mi W of CR 700 E East of Logansport	CASS	LaPorte	A	NONE
U-231	SEG	203+76	203+85	From 0.13 mi S of River Rd (0.45 mi N of Wabash River) to 0.04 mi S of River Road	TIPPECANOE	Crawfordsville	A	NONE
S-311	INT	6+00		SR 311 at SR 60	CLARK	Seymour	X	9611720	2004	INT IMP	3.313
U-030	INT	10+10		S 30 at SR 53	LAKE	LaPorte	A	NONE
U-421	INT	231+90		US 421 at US 20	LA PORTE	LaPorte	X	9703400	2006	INT IMP	2.703
S-046	SEG	83+05	83+13	From 0.67 mi E to 0.75 mi E of the Brown Co Ln	BARTHOLOMEW	Seymour	A	0201349	2006	Partial 3/R	4.388
S-045	SEG	46+42	47+96	From Earl Young Rd to Tunnel Rd Northeast of Bloomington	MONROE	Seymour	A	NONE
S-311	SEG	4+58	5+04	From 0.46 Mi S of then to the Floyd Co Ln	FLOYD	Seymour	X	9611720	2005	INT IMP	3.313
U-030	SEG	9+56	9+77	From 0.54 to 0.33 mi W of SR 53	LAKE	LaPorte	X	0501101	2007	SAFETY	2.500
U-030	SEG	11+92	12+12	From 1.22 mi to 1.42 mi E of I-65	LAKE	LaPorte	A	NONE	2007	HMA
S-037	SEG	143+93	144+23	From 0.32 S to 0.02 mi S of Epler Ave in Indianapolis	MARION	Greenfield	X	0201319&0400678	2006	ATR	11.029
I-064	SEG	123+32	123+61	From the bridge over Spring St to the Kentucky St ln	FLOYD	Seymour	X	0500307	2014	ATR	17.734
U-041	SEG	277+59	277+85	From 121st st to 119th St in Hammond	LAKE	LaPorte	X	0201080	2004	SIG INT	.065
S-135	SEG	132+36	132+52	From Faith St to Main St in Greenwood	JOHNSON	Seymour	A	NONE
U-052	SEG	48+76	48+56	From 0.1 mi N of Kossuth St to Kossuth St in Lafayette	TIPPECANOE	Crawfordsville	A	NONE
S-032	SEG	78+45	78+51	From Elm St to US 31 in Westfield	HAMILTON	Greenfield	A	NONE
U-052	SEG	42+07	42+15	From 0.08 mi W of then to Moorhouse Road	TIPPECANOE	Crawfordsville	A	NONE
S-120	SEG	0+00	0+48	From Middleton Road (Elkhart Corp Ln) then E 0.48 mi	ELKHART	Fort Wayne	A	NONE
S-039	INT	20+46		SR 39 at SR 67 (North Junction)	MORGAN	Seymour	X	0600273&0500844	2009	ATR	3.45
S-007	SEG	17+75	18+48	From 0.04 mi S to 0.69 mi N of CR 330 S Southeast of Mt Vernon	JENNINGS	Seymour	A	NONE
U-031	SEG	124+18	124+24	From 106th St N 0.06 mi in Carmel	HAMILTON	Greenfield	X	0600106	2006	SIGNALS	0.076
U-036	SEG	20+66	20+72	From 0.06 mi W of then to Raceway Road (Marion / Hendricks Co Ln)	HENDRICKS	Crawfordsville	X	0101115	2011	ATR	1.000
S-010	SEG	73+45	73+89	From Linden Rd to US 31 East of Argos	MARSHALL	LaPorte	A	0300496	2003	HMA	1.787
S-055	INT	104+21		SR 55 at US 30	LAKE	LaPorte	A	NONE
U-020	SEG	44+06	44+92	From Meer Rd (CR 600 W) to I-94 Southeast of Michigan City	LA PORTE	LaPorte	A	NONE
U-421	SEG	222+74	222+92	From CR 400 S to CR 375 S North of Westville	LA PORTE	LaPorte	X	0201302	2010	ATR	7.351
I-080	SEG	0+00	0+33	From Illinois State Line E 0.33mi	LAKE	LaPorte	A	Numerous
S-037	SEG	140+53	140+72	From Johnson Co Line N 0.19 mi	MARION	Greenfield	X	0300213	2006	SIG MOD	0.032
U-031	SEG	69+73	69+80	From SR 46 (25th St) N to Haw Creek Blvd	BARTHOLOMEW	Seymour	A	9700230
U-012	SEG	7+61	7+63	From 0.02 mi E of then to Cline Ave	LAKE	LaPorte	A	NONE
U-031	SEG	68+22	68+25	From 0.74 mi N of Base Road to CR 50N SE of Columbus	BARTHOLOMEW	Seymour	X	0014820&0300268	2008	3R/4R	27.662
S-933	SEG	108+50	108+63	From 0.03 mi W of Baker St East to .01 mi E of Taylor St in Mishawaka	ST JOSEPH	LaPorte	A	NONE
S-009	SEG	46+33	46+50	From US 40 (Main St) to Grant St in Greenfield	HANCOCK	Greenfield	A	0400521	2007	HMA	1.261
U-020	SEG	95+00	95+25	From 0.60 mi W to 0.35 mi W of CR 17 Bridge over US 20	ELKHART	Fort Wayne	A	NONE
S-120	SEG	4+84	5+71	From CR 17 E 0.87 mi	ELKHART	Fort Wayne	A	NONE
S-135	SEG	133+92	134+00	From 0.08 mi S then to the Marion Co Ln	JOHNSON	Seymour	A	NONE
S-053	SEG	1+25	1+34	From 0.09 mi S of then to North St (109th St) in Crown Point	LAKE	LaPorte	A	NONE
S-055	SEG	101+19	101+43	From Summit St (105th St) to Birch St in Crown Point	LAKE	LaPorte	A	NONE
U-041	SEG	11+66	11+78	From 0.58 mi S to 0.46 mi S of I-70 (Johnson Ave) in Terre Haute	VIGO	Crawfordsville	A	NONE
I-065	SEG	2+08	2+17	From the Eastern Blvd Bridge over I-65 then N 0.11 mi in Jeffersonville	CLARK	Seymour	A	9018850
U-052	INT	49+91		US 52 at SR 25	TIPPECANOE	Crawfordsville	A	NONE
I-065	SEG	0+00	0+32	From the Ky St Ln to Br over Court/Fourth/Market	CLARK	Seymour	A	9966920	2006	ATR
S-135	SEG	131+82	132+01	From 0.20 mi S then N to Smith Valley Road	JOHNSON	Seymour	A	NONE
U-031	INT	81+39		US 31 at SR 252	JOHNSON	Seymour	A	0300050	2003	Partial 3R	1.089
U-231	SEG	272+10	273+09	From 8th St to 8th Ave in De Motte	JASPER	LaPorte	A	NONE
U-020	SEG	29+64	29+94	From Oak Hill to Waverly Rd in Chesterton	PORTER	LaPorte	A	NONE
U-024	INT	115+13		US 24 at SR 9 (W Jct)	HUNTINGTON	Fort Wayne	A	NONE
S-267	SEG	18+29	18+39	From 0.27 mi S to 0.17 mi S of I-74	HENDRICKS	Crawfordsville	X	9608930	2007	MED CN	4.674
S-038	SEG	72+61	72+82	From US 36/SR 67 E 0.21 mi	MADISON	Greenfield	X	9703270&0500042	2007	INT IMP	1.092
U-040	SEG	7+32	7+53	From 9th St to 11th St of EB one way pair in Terre Haute	VIGO	Crawfordsville	A	NONE
S-057	SEG	49+48	49+53	From Southside Ave/ Troy Rd to Old US 50 in Washington	DAVIESS	Vincennes	A	NONE
I-065	SEG	0+39	0+61	From 0.22 mi S of then to the bridge over 7th St in Jeffersonville	CLARK	Seymour	A	9018850	2006	ATR
U-024	SEG	38+29	39+63	From CR 300 E to CR 400 E W of Monticello	WHITE	LaPorte	X	0500525&0500058	2009	INT IMP	1.340
U-136	SEG	65+65	66+01	From 0.39 mi E of then to Hamstra Rd/ CR 600 E in Brownsburg	HENDRICKS	Crawfordsville	A	NONE
S-003	SEG	195+91	196+35	From SR 8 then N 0.44 mi	NOBLE	Fort Wayne	X	0401187&0300923	2011	INT IMP	0.800
S-056	SEG	117+67	117+90	From Beechwood Ave to Park Ave is Scottsburg	SCOTT	Seymour	A	NONE
U-035	SEG	173+44	174+02	From CR 25 N to SR 8 North of Knox	STARKE	LaPorte	A	NONE
U-041	SEG	270+82	271+07	From 175th St to 173rd St in Hammond	LAKE	LaPorte	A	8665870	2006	Mj PVMT	9.321
U-041	SEG	266+39	266+53	From 45th St to 0.14 mi N in Highland	LAKE	LaPorte	X	9133625	2003	ATR	18.088
U-006	SEG	19+53	19+54	From 0.01 mi W of then to Swanson Rd in Portage	PORTER	LaPorte	X	9299935	2007	ATR	21.742
S-009	SEG	47+22	47+31	From Michigan St to McKenzie Rd in Greenfield	HANCOCK	Greenfield	X	971005A	2004	INT IMP	0.107
U-041	INT	263+18		US 41 at US 30	LAKE	LaPorte	A	NONE
I-065	SEG	229+82	231+57	From 1.68 mi S of then to the Newton Co Ln	JASPER	LaPorte	A	NONE
S-257	SEG	31+26	31+29	From Grove St to Old US 50 in Washington	DAVIESS	Vincennes	A	NONE
U-030	INT	33+74		US 30 at US 421	LA PORTE	LaPorte	X	0500705	2005	3R/4R	2.700
I-265	SEG	0+00	0+32	From I-64 then 0.32 mi E	FLOYD	Seymour	X	041133&0501179	2004	INT MOD	0.775
U-041	SEG	3+46	3+55	From 0.15 mi S of Lincoln Ave to Walnut St in Evansville	VANDERBURGH	Vincennes	A	0501073&9903240	2006	FLASHER	0.009
S-124	SEG	47+72	48+10	From 0.41 mi W to 0.03 mi W of the Wabash River	WELLS	Fort Wayne	A	NONE
U-031	SEG	97+24	97+57	From Tracy Rd (CR 600 W) to 0.13 mi N in New Whiteland	JOHNSON	Seymour	A	NONE
U-031	SEG	158+99	159+01	From 0.07 mi S of Arrow St to Center Rd in Indian Heights	HOWARD	Greenfield	A	0300866&0200094	2004	Partial 3/R	4.359
S-025	SEG	44+95	45+03	From 0.07 mi W to 0.01 mi E of Earl Ave (22nd St) in Lafayette	TIPPECANOE	Crawfordsville	A	0100699	2009	PVMT	9.900
U-040	SEG	100+65	101+66	From CR 400 E to CR 500E E of Greenfield	HANCOCK	Greenfield	A	0400522	2006	HMA	3.077
U-050	SEG	103+20	103+56	From 0.01 mi E of Hancock St to Jackson Park Dr in Seymour	JACKSON	Seymour	A	8923645	2000	ATR	6.745
S-009	SEG	48+29	48+33	From 0.19 mi N of Muskegon Dr to CR 200N in Greenfield	HANCOCK	Greenfield	A	0400521	2006	HMA	1.261
S-110	SEG	12+58	13+24	From 0.07 mi E of US 31 to Michigan Rd	MARSHALL	LaPorte	A	NONE
I-865	SEG	904+34	904+72	From 0.38 mi W of then to I-465	BOONE	Crawfordsville	X	0400882	2018	INT MOD	57.5
S-026	SEG	45+31	46+35	From 0.03 mi E of CR 990E to Clinton County Line	TIPPECANOE	Crawfordsville	A	NONE

ATR	Added Travel Lanes
INT IMP	Intersection Improvement
HMA	Hot Mix Asphalt Resurface
INT MOD	Interchange Modification
FLASHER	Install Flashing Light
3R/4R	Resurface and Bring Roadside up to appropriate 3R/4R Standards
Partial	3/RResurface Minimal Roadside Imprvements
PVMT	Pavement Replacement
MED CN	Construct Median
SIG INT	Signal Interconnect
SIGNALS	Signal Installation
SIG MOD	Signal Modification

REVISED as of Oct 30, 2006

Working Report to Assist INDOT
In Preparation of the 5-percent Report for 2006

REVISED September 8, 2006

Andrew Tarko
August 13, 2006

INTRODUCTION

This working report summarizes the data and measures of safety used to develop lists of segments and intersections in Indiana included in the attached Excel file. This report is meant to help the Indiana Department of Transportation write the 5% report requested by FHWA.

DATA

The data used to develop the lists include: crash data, segment data, and intersection data. The crash data has been extracted from the Indiana State Police Crash Database. The source files are in the text format and include records of all crashes reported on public roads in Indiana in the period of 2003-2005.

The segment data were in the format of a dbf table. It is a part of he Indiana Travel Demand Model and it includes all the segments administered by the Indiana Department of Transportation. These segment data were extracted by INDOT from the Indiana Roadway Inventory Database, supplemented with geographic coordinates and other data, and converted to the dbf format required by the TransCAD. The recent Annual Average Daily Traffic values were included in the segment table.

The intersection data also takes the form of a dbf table and it has been generated with the tool called INSIP (INput to Safety In Planning tool) developed by the Purdue University research team. The segment data were used to identify all the network nodes where at least three state road segments meet. The table of state network nodes includes location, traffic, and geometry data extracted from the road segment table and converted as needed.

The data currently available allow analyzing road segments and intersections administered by INDOT. Inclusion of local roads will be considered for the next year 5% report.

METHOD

The procedure of ranking road locations by level of hazard requires crashes assigned to these locations and proper measures of the safety level.

Assigning crashes to the road network

The crashes reported during the three-year period 2003-2005 obtained from the Indiana State Police were assigned to the Indiana network using the INSIP tool. The original text files were reformatted to fit the TransCAD requirements. A total of 16 tables were created from which the tables consisted of type 10 records were further used. This record type includes all the information needed for the analysis: GIS coordinates of most crashes (app. 80%), names of the streets, time, date, and severity of crashes. In the first step, the INSIP assigned crashes to network nodes based on the geographical proximity. The 250 feet criterion was used according to which any crash that occurred on a road within 250 feet from the center of an intersection is assigned to that intersection. In the next step, the remaining crashes are assigned to the nearest road segments. The outcome of this procedure is the crashes assigned to segments and intersections counted by year and severity. The obtained counts have been attached to the original interaction and segment tables.

The crash assignment procedure also generates a set of tables that link the crash records, with segment and intersection records. They allow checking the detail information about the crashes assigned to specific segments and intersections. These tables were used to verify the correctness of the crash assignment results by displaying on a TransCAD map the crashes assigned to several selected segments and intersections. The check has confirmed good quality of the assignment.

Measures of Safety and Evaluation Criteria

Locations can be ranked (sorted) from two general points of view:

(1) System wide perspective, and

(2) Individual user perspective.

The systems perspective aims to reduce as many crashes as practical in the studied area and to promote the most cost-effective mitigation of hazard. The user perspective aims to reduce the excessive risk faced by individual users, which promotes fairness of the highway system by reducing the difference between risks faced by users at different locations. These two perspectives yield different rank lists.

Annual crash frequency is a system perspective criterion used for intersections. It is a basic measure of crash experience and easy to use as it requires only crash data. The crash frequency is estimated by dividing the number of crashes by the number of years. Selecting locations with a high crash frequency does not consider exposure to risk, i.e. does not take into account volume or vehicle miles traveled (VMT).

Annual crash density is a system perspective criterion used for segments. It provides the normalized annual crash frequency on segments and it is expressed in number of crashes per mile per year.

Crash rate is a user perspective criterion. It is the number of crashes divided by the amount of exposure to risk at the location. For an intersection, a crash rate is calculated as the annual number of crashes per million vehicles entering the intersection. For a roadway segment, a crash rate is calculated as the annual number of crashes per million vehicle-miles traveled (VMT).

Neither a crash frequency nor a crash rate considers uncertainty caused by random variability in the annual number of crashes. Consequently, a high value of crash frequency or crash rate may be caused by randomness and not by high hazard. To incorporate consideration of random crash variability, an index of crash frequency and the index of crash cost are calculated.

Index of crash frequency (I_CF) measures the difference between expected and reported number of crashes divided by the standard deviation of the difference estimate. It combines the system and user perspectives with a stronger emphasis on the system perspective. For example, I_CF = 2 indicates that the number of crashes at the location exceeds the expected number of crashes for that location by two standard deviations. A set of predictive equations is presented Table 1. These equations have been developed for this analysis by fitting the 2003-2005 total crash counts to the segment and intersection data (segment lengths and traffic volumes). Index of crash frequency was calculated using Equation 1.

Equation 1

where :

A	=	number of crashes during Y years (three years here),
a	=	typical crash frequency calculated using Table 1,
Y	=	number of years in analyzed period, in years, and
D	=	over-dispersion parameter taken from Table 1. The smaller factor, the better crash frequency estimate a.

The index of crash frequency can be used in two different ways. In the first method, locations can be ranked using the index of crash frequency. The sorted locations form a priority list for safety reviews, starting with the location for which the evidence of a high crash location is strongest. In the second method, locations can be sorted by other criteria and then the only ones from the top of the list and statistically significant (let say, I_CF >2) are selected.

Index of crash cost (I_CC) measures the difference between expected and estimated crash cost at the location divided by the standard deviation of the difference. It incorporates the crash severity and combines the system and user perspectives with a stronger emphasis on the system perspective. For example, I_CC =2 implies that the crash cost at the location exceeds the expected crash cost for that location by two standard deviations. A set of predictive equations developed for this study by fitting the counts of injury/fatal and PDO crashes to the segment and intersection data are presented in Table 2. Index of crash cost is calculated using Equation 2.

Equation 2

where:

CPD	=	average cost of PDO crash, here $6,500,
CIF	=	average cost of I/F crash, here $65,000,
PD	=	number of PDO crashes during Y years,
IF	=	number of I/F crashes during Y years,
aPD	=	typical PDO crash frequency, in PDO crashes per year,
aIF	=	I/F crash frequency, in, I/F crashes per year,
Y	=	number of years in analyzed period, in years,
DPD	=	over-dispersion parameter for PDO crashes, and
DIF	=	over-dispersion parameter for I/F crashes.

RESULTS

The Excel file provided with this report includes an Intersection spreadsheet with 1085 intersections and a Segments spreadsheet with 19,262 inventory segments. The color-coded columns have names that are self-explaining or explained in the above part of this report. The blue color indicates location and ID information needed to display the results on an electronic GIS map or to find the segment or intersection on a map. The yellow color indicates all the input values used to calculate the results. The orange color indicates the results. The remaining white columns contain wealth of additional information about segments and intersections. The variable (column) names for segments are explained in the manual for the Travel Demand Model. The variable (column) names for intersections are in the draft final report for the JTRP project on safety-conscious planning. These descriptions can be provided upon request if needed.

The tables can be sorted by any of the columns containing measures of safety as introduced in the previous section of the report. Other can serve as secondary measures. If using measures other than Index of Crash Frequency and Index of Crash Cost, then the segment and intersection lists have to be kept apart and sorted individually. The value of the Indices of Crash Frequency and Cost give an indication of the statistical significance that a location experiences excessive hazard.

Another alternative is to combine the segment and intersection lists and sort the combined list by Index of Crash Cost. Other measures such as crash rate can be used to make a final selection.

In any case, the segments that are too short may be removed.

Table 1 Safety performance functions

Facility	Safety Performance Functions	Over-dispersion parameter (D)
Intersections	a = 3.17.Q10.403Q20.416	0.460
Rural non-interstate segments	a = 2.291LQ0.809	0.430
Rural interstate segments	a = 0.557LQ0.978	0.086
Urban non-interstate segments	a = 4.747LQ0.860	0.893
Urban interstate segments	a = 0.230LQ1.350	0.225
a = typical crash frequency in Indiana , in crashes per year, Q = AADT entering the intersection or along the road segment, in thousand vehicles per day, D = over-dispersion parameter, and L = road segment length, in miles.

Table 2 Safety performance functions including severity

Facility	Safety Performance Functions	Over-dispersion parameter
Intersections	aIF= 0.913.Q10.271.Q20.456	0.257
	aPDO= 0.238.Q10.415.Q20.421	0.434
Rural non-interstate segments	aIF= 0.462LQ0.825	0.120
	aPDO= 1.747LQ0.793	0.370
Rural interstate segments	aIF= 0.065LQ1.096	0.005
	aPDO= 0.506LQ0.950	0.082
Urban non-interstate segments	aIF= 0.959LQ0.886	0.565
	aPDO= 3.394LQ0.870	0.863
Urban interstate segments	aIF= 0.913LQ0.456	0.257
	aPDO= 0.181LQ1.358	0.218
aPD = typical PDO crash frequency, in PDO crashes per year, aIF = typical I/F crash frequency, in I/F crashes per year, Q = AADT entering an intersection or along the road segment, in thousand veh/day, L = road segment length, in miles, and D = over-dispersion parameter.

CLOSURE

The present report does not include local roads due to the lack of sufficiently complete representation of Indiana road network. A progress is being made to include local roads. An effort is planned to incorporate local roads in the next year report. Another issue that needs to be addressed is the frequent lack of data for local roads. It may require modification of the method used this year.