November 12, 2014
- MEMORANDUM FOR:
- REGIONAL ADMINISTRATORS
- DOROTHY DOUGHERTY
Deputy Assistant Secretary
- FRANCES YEBESI
Acting Director, Directorate of Evaluation and Analysis
- Establishment-Targeting Lists for Emphasis Programs
Various OSHA directives governing National, Regional, and Local Emphasis Programs explain how an inspection list must be compiled, ordered, and randomized. This memorandum provides additional instruction on generating and randomizing these establishment-targeting lists. If the instructions in this memorandum conflict with the instructions contained in an Emphasis Program directive, the Area Office must attach this memorandum to the Emphasis Program's targeting list and include it in any request for a warrant authorizing an inspection to be conducted pursuant to the Program.
The April 21, 2014, memorandum Generation and Randomization of Inspection Targeting Lists is rescinded.
List Generation and Randomization
For National, Regional and Local Emphasis Programs, the National Office's Office of Statistical Analysis (OSA) will prepare a Master List for each Area Office, using OSA's most current establishment file. OSA will apply the industry and establishment-size criteria prescribed by the particular Emphasis Program to select all eligible establishments. OSA will assign a random number to each establishment on the OSA-generated Master List and provide the Master List to the Area Office, sorted in random number order.
OSA updates the establishment file every six months.
Area Offices may add establishments to the OSA-generated Master List, based on sources that may include, but are not limited to: (a) commercial directories; (b) telephone listings; (c) local knowledge establishments, derived from previous OSHA inspection history; and (d) information from other government agencies, such as the local Health Department. The Area Office must retain documentation of every addition made to the Master List and, prior to use, the revised List must be re-randomized -- either by OSA or by the Area Office using the RANDBETWEEN function in Microsoft Excel.
Outside of manufacturing, the OSA establishment file is limited to certain industries and to establishments with 10 or more employees and, consequently, may not be appropriate for generating some Local Emphasis Programs. If a Local Emphasis Program or a Regional Emphasis Program covers industries not included in the OSA establishment file, the Area Office may generate its own master list (see number 2 above for potential sources). The Area Office may provide the list to the National Office for assignment of random numbers or the Area Office may assign random numbers to the list using the RANDBETWEEN function in Microsoft Excel.
The Office of Statistical Analysis (OSA) has automated the process of compiling, randomizing and delivering NEP and LEP inspection targeting lists (i.e. Master Lists) to the field offices. Area Offices can access and download their Master Lists from the ListGen webpage. The selection criteria for the Agency's National Emphasis Programs are preloaded within the application. Area Offices can load the selection criteria for their Local Emphasis Programs.
The ListGen application uses Oracle's "DBMS __ Random'' tool to assign random numbers to the list.
An Area Office should generate only one list per emphasis program for a given file. The application selects and returns ALL establishments meeting the selection criteria (it does not sample). Multiple lists generated from the same file would return the same establishments. The file date is displayed within the system and on the generated reports and the application will provide a warning if a second list for a given program is generated from the same file.
The Area Office may print and export the Master List. Once created, Master Lists remain available within the system.
[Paragraph removed. For internal OSHA use only.]
NAICS codes are updated and modified every five years. This introduces the potential of having different sets of NAlCS codes for the scope of the emphasis program and the make-up of the establishment file. For example, the current establishment file used by OSA is based on 2012 NAICS codes, while several emphasis programs use 2007 NAICS codes to define their industry scope. When this issue arises, the Area Office should use the U.S. Census concordance tables at http://www.census.gov/cgi-bin/sssd/naics/naicsrch?chart=2012 to identify any relevant differences between the two coding systems. The appropriate establishment file codes will then be used to generate the master list. Any variance from the list of industries included in the directive should be documented and attached to the inspection targeting lists.
A "cycle" is a subset of establishments from the Master List and allows the Area Office to use its enforcement resources more efficiently: rearranging establishments' order within a cycle is reasonable, because it furthers OSHA's legitimate goal of efficient allocation of inspection resources.
The Area Office can schedule inspections in the following manners:
1) If the Area Office intends to inspect the entire Master List, it can schedule the inspections in any order. If the Area Office uses this method, it must account (through inspections and deletions) for the entire list. The Area Office cannot inspect off a new list until the entire previous list is completed.
2) The Area Office can inspect the facilities in the random number order provided. If the Area Office uses this method it does not need to complete the entire Master List.
3) The Area Office can create cycles by choosing a set number of establishments from the Master List, sorted in random number order. For example, the Area Office can select the first ten establishments on the Master List, sorted by ascending random number as a first cycle of 10 establishments. The Area Office can then inspect those establishments in any order, but must finish the cycle prior to beginning a second cycle. Once the first cycle is complete, the Area Office can inspect the next ten establishments on the randomized Master List as its second cycle.
Most NEPs and LEPs are based on industry selection criteria. Selection of establishments relies on the industry coding provided in the universe establishment file (currently the D&B file). The industry coding of establishments in the universe file reflects reasonable groupings of particular establishments within appropriate categories based on information available to the publisher of the file. This also applies to other file data elements such as employment size, physical address, or operating status (for example the facility may be recently closed or out-of-business). As part of an Area Office's pre-inspection research, a thorough web-search of the facility may be conducted to determine if there is viable evidence showing the establishment falls outside of the emphasis program's selection criteria. Based on evidence found, the Area Office may delete the establishment from the targeting list and must provide documentation of the reason for the deletion and the source of the information relied upon. This documentation should be kept as an attachment to the targeting list.
New LEPs. REPs and NEPs
As new LEPs, REPs, and NEPs are developed, the guidance of this memorandum must be inserted in the section of the directive that prescribes how to generate and randomize inspection-targeting lists. OSHA's National Office will update Appendix C of CPL 02-00-025, Scheduling System for Programmed Inspections (January 4, 1995) to include this guidance.