RBG-25104, Responds to 860627 Request for Addl Info Re Facility Spds. Human Factors Engineering Evaluation Performed on Emergency Response Info Sys,Spds & Digital Radiation Monitoring Sys. Detailed Description of Evaluation Methodology Encl

Responds to 860627 Request for Addl Info Re Facility Spds. Human Factors Engineering Evaluation Performed on Emergency Response Info Sys,Spds & Digital Radiation Monitoring Sys. Detailed Description of Evaluation Methodology Encl

ML20207D347
Person / Time
Site:	River Bend
Issue date:	12/17/1986
From:	Booker J GULF STATES UTILITIES CO.
To:	Harold Denton Office of Nuclear Reactor Regulation
References
RBG-25104, NUDOCS 8612300380
Download: ML20207D347 (13)

Text

.--

GULF STATES UTIL WTIES C'OMPAAY RIVER oEN3 STATION POST OFFICE ROX 220 ST. FRANCISVILLE. LOutSIANA 70775 ARE A CODE 504 635-6094 346 8651 December 17, 1986 RBG- 25104 File Nos. G9.5, G9.33.4 Mr. Harold R. Denton, Director Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Dear Mr. Denton:

River Bend Station - Unit 1 Docket No. 50-458 This letter is Gulf States Utilities Company's (GSU) response to a request for additional information dated June 27, 1986 regarding the River Bend Station Safety Parameter Display System (SPDS). Attachment 1 addresses those concerns raised by the Staff to complete a confirmatory review of the SPDS. An update of implementation status of the referenced human engineering descrepancies (HED's) will be provided during July 1987.

In addition, supplemental information to GSU's letter dated September 29, 1986 concerning the selection of SPDS monitored variables is being provided in Attachment 2.

If additional information is required, please contact Mr. Rick King at (504) 635-6995.

Sincerely, o M J. E. Booker Manager-River Bend Oversight River Bend Nuclear Group JEB/RHG/hk/je Attachment b

O I

/

B512300380 B61217 O PDR ADOCK 05000458 AD P PDR F

\

ATTACHMENT 1 River Dend Station (RBS)

SPDS and DRMS Supplemental Information 1.0. GSU has not validated the suitability of the Emergency Response Information System (ERIS) Safety Parameter Display l System (SPDS) (ERIS and the Digital Radiation Monitoring System (DRMS)) as integrated with the RBS control room, Emergency Operations Procedures (EOPs), and operator training.

A. GSU performed a human factors engineering evaluation of the ERIS SPDS and DRMS. A detailed description of the evaluation methodology is provided in Enclosure 1.

2.0. GSU has not evaluated the applicability to RBS of human factors comments identified- during the generic GE validation program conducted at the Perry plant.

A. During the human factors engineering evaluation of the ERIS SPDS and DRMS, the applicability of the human factors comments identified during the generic GE validation program conducted at the Perry Plant were considered for inclusion as HED's, if applicable.

3.0. GSU has not conducted a formal verification and validation program for the DRMS.

A. The RBS DRMS software development was conducted in accordance with the GA Technologies, Inc. Verification and Validation (V&V) program. A meeting was held between GSU and the NRC staff on August 7-9, 1984 which discussed DRMS V&V. A NRC meeting summary, dated October 17, 1984 from Edward J. Weinkam, III, discussed, on page 7, that confirmation by GSU that its DRMS V&V program is identical to the program applied on the Seabrook DRMS wculd be acceptable to the staff. The NRC staff has reviewed and l accepted the GA V&V software methodlogy for the Seabrook l DRMS V&V software methodology developed and controlled by l GA Technologies.

The software methodology used for the development of the Seabrook DRMS equipment is the same as that used in developing and controlling the RBS DRMS software. RBS has not employed design modifications or changes to its DRMS software without GA Technologies, Inc. review and approval in accordance with GA's approved V&V software methodology.

GSU will model its plant specific DRMS software V&V program similar to the GA V&V program and is scheduled for implementation by the end of the first refueling outage.

Future DRMS software changes prior to the end of the first refueling outage will be processed utilizing the GA DRMS

ATTACHMENT 1 (cont'd.)

software V&V approved methodology. Therefore, the current RBS DRMS V&V software methodology is considered acceptable.

4.0. The ERIS keyboard for the SPDS contains unused keys and function keys that are _not assigned and easily identifiable.

A. HED Nos. 882 and 904 have been initiated to address the subject concern. EEAR No. 86-E0103 is being processed on the ERIS SPDS keyboard to evaluate unused keys and function keys that are not assigned and easily identifiable.

5 0. Selection of displays on the SPDS (ERIS) is cumbersome and awkward. The operator must continuously return to a menu to ascertain what must be done to display other desired screens.

A. HED No. 894 has been initiated to determine the solution (See Enclosure 1). EEAR No. E0106 is in place to review if a design change is the correct solution to the subject concern.

6 Q. The SPDS (ERIS) presently contains many dummy screens that were used by designers during display development. These dummy screens tend to cause operator confusion and slow down operator access time.

A. Dummy. displays have been removed from the ERIS SPDS host computer, making them inaccessible from the SPDS graphic display terminals.

7.Q. The indication of FRIS system failure is not readily apparent. During the audit, operators informed the NRC Audit Team that-they have frequently, inadvertently used an inoperable SPDS during simulator training.

A. HED No. 902 is being processed to determine the correct action to provide conspicious indication of system failure.

8 0. The DRMS problem which allows masking of alarm conditions should be corrected.

A. Modification Request No. 86-0362 has been initiated to review and determine corrective action for the masking of incoming alarms on the DRMS.

9.Q. Inconsistency between color coding (i.e. DRMS blue vs ERIS SPDS blue) used on the ERIS and DRMS.

ATTACHMENT 1 (cont'd.)

A. HED No. 919 was initiated to review color coding inconsistency between the ERIS and DRMS. The cyan color as used on the ERIS SPDS displays will continue to be used to prevent possible misinterpretation of validated data as implying a " safe" condition, which is color coded using green. The difference between the cyan color on the ERIS SPDS and DRMS displays is not considered critical since the primary colors used on the systems agree in meaning (i.e.

green' indicates a safe condition, yellow an alert condition and red an alarm condition). Therefore, no additional action is planned on this recommendation.

10.0. Antipodal location of the DRMS and ERIS displays in the Main Control Room.

A. The location of the displays has not been cited by operators as presenting a problem. In addition, this item was not identified during the human factors evaluation of the ERIS SPDS and DRMS as a discrepancy. Therefore, no change is planned for the location of the termination.

11.0. All control functions should be removed from EOF /noncontrol room SPDS/ERIS/DRMS terminals. If not removed, acceptable safeguards need to be implemented.

A. The URMS software has been modified to allow control functions to be performed only from the RBS Main Control Room.

12 0. The complex color coding system used on ERIS and the lack of a color code menu could cause the operator difficulty or confusion in using the SPDS.

A. HED No. 889 has been initiated to address the seven different colors used to coding five distinct kinds on information on ERIS SPDS. A hardware / software change request (No.86-115) has been initiated to provide a display for explaining the color codes used on sections of the ERIS SPDS displays.

J

ATIACINENT 2 SPDS Variable Selection Information

1. Radiation Monitor Information On SPDS In GSU's letter dated May 15, 1985, fran J. E. Booker to Harold R. Denton (RBG-21018) the critical safety function radioactivity control monitored variables were identified in Table 1. In addition, Table 1 to the staff's letter dated June 27, 1986, contained the monitored variables for radioactivity control.

2. Containment /Drywell Hydrogen Concentration Currently, GSU is participating in the Mark III Hydrogen Control Owners' Group (HCOG) generic effort to develop a BWR Mark III combustible gas control energency procedure guideline. Once approved by the staff, the hydrogen control guideline will be reviewed for incorporation into the existing River Bend Station (RBS) Emergency Operating Procedures (EOPs) . Upon ompletion of developing a RBS specific hydrogen control E0P, GSU will consider the need to inplanent a containment /drywell hydrogen concentration variable on its RBS SPDS displays. 'Ihe current RBS design arploys safety related hydrogen analyzers with safety related indication provided in the main control romi for hydrogen control.

ENCLOSURE 1 l

r 9

HUMAN FACTORS EVALUATION OF THE RIVER BEND STATION I SAFETY PARAMETER DISPLAY SYSTEM This report documents the methodology and findings of the human factors engineering evaluation of the Emergency Response Information System (ERIS) and the Digital Radiation Monitoring System (DRMS) , which together make up the River Bend _ Station (RBS) Safety Parameter Display System (SPDS).

The General Physics Corporation (GPC) was contracted to assist GSU in developing and implementing this human factors evaluation.

Mr. Arthur Beare, PhD, a human factors specialist with GPC, worked with GSU Plant Staff to perform the tasks outlined in this report.

The focus of the evaluation was on hardware and display characteristics that affect the ability of the operators to use the systems efficiently. The evaluation employed a multifaceted approach that included:

Detailed evaluation of representative displays against guidelines from NUREG-0700 An Operability Verification including observation and debriefing of crews using ERIS and DRMS while responding to simulated transients in the RBS simulator Questionnaires and interviews' to elicit operator feedback about various aspects of both systems Extensive discussions with the ERIS and DRMS system engineers Overall evaluation of ERIS and DRMS against checklists incorporating guidelines from NUREG-0700 and Appendix A to SRP Section 18.2 of NUREG-0800 Methods The methodology employed in each part of the evaluation is described below. Discussion of findings is deferred until the end of this report. -

Display Evaluations All "SPDS" displays available on ERIS cnd several representatives of each type of display available on the DRMS were reviewed.

Checklists from sections of NUREG-0700 dealing with data presentation format (secti6n 6.7.2.4), screen layout and structuring (6.7.2.5) and graphic coding and highlighting (6.7.2.7) were completed for nine representative ERIS displays and 10 representative DRMS displays. Discrepancies noted were evaluated on an item-by-item basis.

Operability Verification A dynamic operability verification was performed in the RBS simulator to determine if control room personnel would have any difficulties in using the ERIS and DRMS to monitor the plant's status under simulated emergency conditions. The evaluation was conducted in conjunction with simulator training sessions performed as part of the licensed operator requalification training program.

The six scenarios were:

1. Total Loss of High Pressure Makeup

2. Loss of Drywell Cooling

3. Steam Leak in Main Steam Tunnel

4. Loss of Normal and Emergency Feedwater

5. Fuel Cladding Failure

6. Anticipated Transient Without Scram (ATWS)

The scenarios were designed to exercise the RBS Emergency Cperating Procedures and ranged in severity from very mild to quite severe. ,

Each scenario was performed by each of two crews consisting of licensed SBOs and R0s and assisted by a Shift Technical Advisor.

The evaluation of the exercises was based on observations of ERIS and DRMS use and formal debriefings of SPDS users using a protocol based on " Human Factors Review Guidelines for the Safety Parameter Display System", Appendix A to SRP Section 18.2 of NUREG-0800. The items on this protocol focus on issues of system usability and the debriefings provided a means for the users to evaluate ERIS and DRMS in light of their experiences with the systems during the simulator scenarios.

Questionnaires and Interviews A questionnaire to obtain opinions about various aspects of ERIS and DRMS use was distributed to licensed operators in February.

The questionnaire had four sections:

l 2

l i

1

1. Questions for assessing the respondents familiarity with j

ERIS and DRMS.

i

2. A series of true-false questions relating to aspects of ERIS that had been identified as potential problem areas in earlier evaluations (one of ERIS conducted at the Perry simulator and one of the RBS systems conducted by NRC Staff) .

l 3. ~ Scales for rating ERIS display characteristics.

4. Open-ended questions asking the respondents to identify characteristics of ERIS and DRMS that they felt should be

?

changed.

Completed questionnaires were returned by seven SROs. The

! confirm the existence of Human Engineering responses tended to Discrepancies (HEDs) identified by the earlier evaluations, and

, the open-ended questions identified a number of additional areas of concern.

Interviews were conducted with five of the SROs who returned the questionnaire. The interview protocol included questions on the case of determining if the systems were operating properly, the use of color (including the potentially conflicting color codes for ERIS and DRMS) , display selection, the adequacy of training in the use of the systems, and a number of questions relating to problems identified by earlier reviewers or in responses to the questionnaire.

Overall Evaluation l The overall evaluation of the two systems comprising the RBS SPDS l was performed using copies This of a checklist entitled, " Human Factors SPDS Guidelines". checklist is comprised of items taken from several sections of NUREG-0700 and items drawn from Guidelines for the Safety Parameter l the " Human Factors Review Display System (SPDS)" presented as Appendix A to SRP Section 18.2 of NUREG-0800. Information collected from the detailed display reviews, operability Verification debriefings, questionnaires, interviews, and direct measurements (e.g. ,

. display presentation times, display luminance) was used to address each item on the checklists.

1 Maior Findings and Recommendations Both the ERIS and the DRMS are currently operational at River Bend Station and are used daily by control room personnel. While l

1 both systems have a number of good features, the present evaluation has identified or confirmed thesystems existence of human engineering discrepancies in both as currently implemented. This section summarizes these HEDs with corrective j actions recommended.

3 i

.m,___ - _ - _ - _ , . - . . ,

1) The present system of display selection on ERIS requires the user to return to the Format Menu whenever he wants to select a display *. GPC recommended that a special-purpose keyboard. allowing one-button selection of all SPDS displays be considered. In the interim, simply assigning the most useful SPDS displays to the currently available function keys and labeling each key with the name of the display it selects will significantly streamline the display selection process.

2) The color coding scheme used on the ERIS displays is complicated.and not understood by all users. It was recommended that a display explaining the color codes used with various display elements be added to the system. In addition, three changes to the color coding were recommended:

A) To avoid a potential conflict with the color codes on the DRMS, where cyan is used to code potentially

" bad" data *, it was recommended that green rather than cyan be used to indicate validated data on ERIS.

B) To improve consistency with the color coding of limit tags, it was recommended that the Status Windows be green instead of blue when all parameters are within normal limits.

C) To provide a more consp'icuous indication of changes from " active" to " inactive" status for the systems represented in the Systems Status Matrix, the borders should turn from green to red rather than simply reverting to blue (red borders would change to blue [

upon acknowledgement from the operator) .

3) The time required to draw a complete trend line on the ERIS single-parameter trend displays can exceed one

minute. It was recommended that the time required to draw the trend lines be reduced as much as possible, and l, that the programming be altered to allow rescaling before l the trend line is completed.

l 1 I 4) Although all ERIS displays include a clock that updates at roughly two-second intervals, a more conspicuous '

indication of system failure is needed*. It was recommended that all failures be signaled by an audible alarm, and those failures not resulting in blanking of the display should result in conspicuous color changes.

4

• This and other items marked by an asterisk were noted in the NRC staff's safety evaluation of the River Bend SPDS.

4 l

5) The luminance of red on the ERIS CRT is 'significantly-

less than 'that recommended in NUREG-0700. Since yellow and red serve as visual. alarms, .the CRTs should be adjusted to provide greater luminance'for red.

6) Alarm priorities should be changed so that radiation alarms and alerts have the highest priorities on the DRMS 1

miniature grids * (an MR to effect this change had been submitted by the DRMS System Engineer).

7) The' graphic representations of radiation values on 'the l DRMS Trend and Group displays only display one decade above the alarm setpoint, which severely limits the i usefulness of these displays for depicting high radiation

, levels. It was recommended that the graphic scales on these displays 'be extended several decades above the alarm setpoints.

8) Licensed operators had received littleuformal instruction on the ERIS and many of those interviewed did not

! understand all the features of the system. Additional

~ training is needed to provide a reasonably detailed understanding of the system.

Detailed operators manuals are not

9) available in the

. control room for either ERIS or DRMS. It was recommended that such manuals be provided. -

Evaluation of Findings To determine what corrective actions would be implemented, GSU created a task force of plant staff personnel who were familiar 3

with the design and operation of the systems. This group reviewed each HED and performed the following tasks:

, 1) Established a criticality rating for each HED using the i table shown on the next page. Since the systems are not safety-related, all ratings were either 3 or 4.

2) Reviewed each of GPC's recommended corrective actions to 4

determine their feasibility, whether their implementation l was desirable, their cost effectiveness, and if there j were alternative solutions.

3) An implementation schedule for corrective actions was developed for each HED. Due to the engineering research that will be required to determine the feasibility of some of the proposed changes, actual completion dates

could not be determined. Instead, a time frame for completion was allotted after the engineering evaluation is completed.

1

- 5 i

)

i RD PRIORITIZATION MMRIX Error Potential Documented and/or Low Potential High Potential Error Error si

, tz T I E 11 3

en I D I

m i

uli

~m lie iv l y I

f 1

I I

l l

~

. l3 L .

~

The results of this evaluation are documented on the HEDs contained in the next section.

Human Engineering Discrepancies Forty-one (41) HEDs were written which represent all of the findings from this human factors evaluation. They are formatted similarly to the Detailed Control Room Design Review (DCRDR) HEDs so that the same computer database cou1d be used. Please note that the HED numbers are reference numbers only. The numbering begins at 881 to follow in progression with previously documented HEDs.

. .