[labnetwork] More on buddy system and a labnetwork suggestion / Fwd: 24/7 operations

Tue May 31 13:38:06 EDT 2011

Dear lab network,

Great forum.  Thanks to all for important feedback which I have compiled in the thread below.

Here at the Utah nanofab, we run 24/7 with training aimed at proper off-shift use of tools, and we work to create an enforceable buddy system, and build systems that make it easier for labmembers to adhere to the off-shift buddy system.  Our intent is: not having a safety buddy is not an option.

Among the tools we use to make it easier to adhere to the buddy system (and to silence complaints from faculty that the buddy system requirement is too onerous!) include:
a buddy system online calendar, buddy system mailing list, and even buddies for hire (trained undergrads willing to sit in the lab (doing homework) and be a buddy, for pay.  No one has used this last tool since it was implemented a year ago, but it has helped silence the faculty critics)
http://fab.eng.utah.edu/buddy-tools

Outside the fab we have a monitor indicating who is inside (buddies are not required in the adjacent SEM/XPS lab or in the packaging lab).
http://fab.eng.utah.edu/index/about-us/Home/lab-members/userinlab

And we use the combination of card-key records (we have our own system, and so have real-time access) and video monitoring (including at wetbenches) to enforce the rule as others have indicated.  

Recently we began generating a monthly report of buddy system violations, sorted by research group, and send the relevant violations to the faculty PI along with their user fee invoice, so that he or she is aware of what is going on with their own group.  We do not necessarily associate (yet) disciplinary measures with this report, as we wish to initially just raise the visibility of the need for adherence (and the fact that we are paying attention) first.

We do not yet have a culture wherein the labmembers themselves feel a sufficient sense of ownership that they are generally willing to either confront a safety violator in the lab, or else report violations.  But it is part of our formal training, that "someone else's safety violation certainly affects you in a number of ways, including..."  Such a culture is an ideal that we are striving for: self-monitoring & self-correction among the labmember community.  In order to achieve this sense of ownership, belonging, responsibility & accountability, we have had our student representatives assist us by actually writing the policies and being involved in scheduled and weekly lab cleans.
http://fab.eng.utah.edu/uploads/pdfs/Nanofab_User_Policy%20Aug_2010.pdf

As we prepare to move into our new facility, we will be making additional improvements to the buddy system:
(1) by implementation of CORAL and interlock boxes on each of the tools (now in progress), we will be able to begin billing based on time spent on a tool, rather than time spent in the lab.  Our interest in the buddy system is a key driver for this administrative change.  We hope that by so doing, a researcher will be able to more easily find a colleague to be a buddy (or associate in the same research group, perhaps), since that buddy will not have to pay to be in the lab.

(2) our new facility includes a clean conference room attached to the fab and accessible in the bunny suit as well as from a pedestrian entrance (think of the negotiation table between N & S Korea, entered on either side with a symbolic barrier to crossing between.  Since ours is not complete yet, here is a link to the Korean table!).  
http://www.traveladventures.org/continents/asia/panmunjom2.shtml
One of the functions of our clean conference room at night will be for buddies not working on the fab tools to have a place to sit in a bunny suit and work on reports or mask designs or device simulations.

(3) Carding out of the lab, we will be implementing an annunciator to indicate if someone is about to exit, and leave someone else stranded without a buddy.  Clearly, the policies and software methods for allowing someone to enter alone (needs a buddy to also enter within a prescribed period) or exiting (and potentially stranding someone) requires special attention, and we are working on how to implement this presently.

ASIDE: As a result of the recent informative LABNETWORK thread on TMAH, we are implementing changes in our lab.  My recent question posed to the LABNETWORK on alarm response protocols was also very helpful in providing us much useful information. 

I suggest that in addition to the UGIM meetings held once per year and which describe these types of management topics, Perhaps we should also have some forum in the "off-years" to get into detail regarding the soft administrative systems, and provide a means of sharing not only best practices, but also code for the soft systems themselves.  For example we are implementing CORAL with home-designed interlock boxes with off-the-shelf components, and are quite willing to share our design and the associated CORAL software modules.  We are currently in the planning/design stage of our new prox-card entry and tool enable system with associated buddy system features.  Others have implemented add-ons to CORAL, or perhaps generated their own soft admin systems with modules that can be ported to others.  Is it worthwhile attempting something such as this?  We at Utah would be happy to host the first, unless someone else would like to do it.

Thank you all,

--Ian

********************************************
Ian R. Harvey, Ph.D.
Research Associate Professor
Department of Mechanical Engineering
Adjunct Associate Professor
Department of Electrical & Computer Engineering
2232 MEB

Associate Director, Utah nanofab
College of Engineering / University of Utah

mail to suite 2110 MEB, 50 S. Central Campus Drive
Salt Lake City, Utah   84112-9011
801/585-6162 (voicemail)
801/581-5676 (lab main number)
www.nanofab.utah.edu

Begin forwarded message:

From: "Hathaway, Malcolm" <hathaway at cns.fas.harvard.edu>
Date: May 31, 2011 7:07:41 AM MDT
To: "labnetwork at mtl.mit.edu" <labnetwork at mtl.mit.edu>
Subject: Re: [labnetwork] 24/7 operations

Hi all,

This is Mac Hathaway, at Harvard CNS.  We have protocols in place similar to those described by John, with the exception that our "toxic" gases (flammables and anything monitored with sensors, Cl2, SiH4, BCl3, etc) are only enabled from 6 AM to 8 PM.  We have had discussions about whether this restriction is still needed, given the maturity of our toxic gas monitoring, but 6 AM-8 PM is the current scheme.  Everything else in the cleanroom is physically accessible 24/7 (most notably, from a safety standpoint, wetbenches).

One thing that is a little different here is that we have an ERT (Emergency Response Team), similar to what exists in many commercial fabs, (not typical for academic sites, as I understand it).  The ERT consists of most of our cleanroom staff.  We meet monthly for training (and pizza), and we are all on call in the event of an emergency in the cleanroom.

We have the buddy rule for working with hazardous materials, with a strong emphasis on the need for a buddy for any work during "off-hours", but as John pointed out, it's not something that lends itself to absolute compliance.  With regard to "encouraging compliance", how do people feel about the use of video cameras on the wetbenches?  I understand that some locations are using them;  has wetbench user "behavior" improved when cameras are in use?  Obviously, 24-hour monitoring is not practical, but does the "someone is watching" effect yield positive results?

Mac Hathaway
Senior Process Engineer
Harvard CNS

Begin forwarded message:

From: "Luciani, Vincent" <vincent.luciani at nist.gov>
Date: May 31, 2011 7:29:28 AM MDT
To: "'Matthieu Nannini, Dr.'" <matthieu.nannini at mcgill.ca>, "labnetwork at mtl.mit.edu" <labnetwork at mtl.mit.edu>
Subject: Re: [labnetwork] 24/7 operations

Hello Matt,

Here at the CNST NanoFab (http://www.nist.gov/cnst/index.cfm), we are staffed from 7 AM to midnight,  M-F.  We use card key access control system and Coral.  All users are locked out after hours and we program their afterhours access into the system after their request for after hours access is approved.

·         Advance notification, the day before is fine.  That way we can avoid any tool repair or facilities conflicts. The advance notification also helps me help others to find a buddy if needed.
o   A loose, casual buddy agreement between users, leading to one user leaving before another or one working in their office while the other was in the lab was a common failure mode at first. So, we require verbal or email confirmation from both people that they are committed to being each other's buddy at all times.
·         Once approved, they can use all tools/processes.
·         We use deterrence to discourage cheating, because it will happen.  I randomly check Coral records and our CCTV video tapes and suspend/revoke cleanroom privileges for violators (from a 2 week suspension to full revocation depending on the degree of fraud/abuse).  Word travels fast when this happens and is a strong deterrent.
·         Some tools outside the cleanroom, like the FIB or AFM, can be used without a buddy after I confirm with whoever trained them that they have the required proficiency.
·         The NanoFab staff is notified automatically via Blackberry about any gas or fire alarm 24/7.

Let me know how it goes for you.

Vince

Vincent K. Luciani
NanoFab Manager
Center for Nanoscale Science and Technology
National Institute of Standards and Technology
100 Bureau Drive, MS 6201
Gaithersburg, MD 20899-6200 USA
+1-301-975-2886
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.jpg
Type: image/jpeg
Size: 2876 bytes
Desc: not available
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20110531/dc68dc6d/attachment.jpg>
-------------- next part --------------

******************************************************************************

Begin forwarded message:

From: Dan Woodie <daniel.woodie at cornell.edu>
Date: May 31, 2011 8:13:06 AM MDT
To: "labnetwork at mtl.mit.edu" <labnetwork at mtl.mit.edu>
Subject: Re: [labnetwork] 24/7 operations

Matthieu,

Like the Stanford Facility, the CNF is open 24/7 with staffing only the standard work week (7 AM ~ 6 PM, M-F). Once users complete our orientation process (~ 7 hours spread over 1.5 days) they are given full 24/7 access to the facility. Note, this does not include training on tools, only general lab orientation and chemical hood training. U.S. law assigns a greater responsibility to oversee undergraduate students versus graduate students, so we restrict undergraduates to only be able to work during normal staff hours (8 AM - 6 PM, M-F). They can go through a staged process of experience in the lab and staff review to gain full lab access over time.

Unlike the Stanford model, we do have access to the card access system, and we have large screen displays posted around and in the cleanroom listing who is currently in the facility. This can help users determine if they are working alone or not. We require a 'buddy' to be present for a few tools and our high contact hazard chemical hoods, but only evenings and weekends when staff are not on duty. For the chemical hoods we have a sign in process required all the time, with the buddy needing to sign in with the user after hours. 

Similar to Stanford we restrict the users from accessing any of the building or tool utilities and severely limit what operations they can process on tools. Emergency phones are located throughout the lab which will automatically connect to the Cornell Police Dispatch if lifted, and they will investigate even if the phone is just lifted and then hung up. Additionally we have a comprehensive toxic gas monitoring and control system present to alarm and notify the campus police dispatch and staff remotely in case of any gas detections. Cornell EH&S also full 24/7 coverage for chemical spills, injuries, etc. that can respond very quickly and is somewhat knowledgeable on our hazards and safety systems. They can respond to gas detection events, typically with over the phone support from our staff.

Lastly, we have video cameras covering all tools and chemical hoods in the facility, with a digital recorder that only captures when motion is sensed. We don't use it too often, but it has been very valuable as a deterrent. Before we implemented it, a subculture of users developed who worked at night, thinking they could skirt the rules to do things faster or different. Once the cameras went up and we handed out some suspensions from observed activity, that went dramatically down. Now, we get less argument from violators when we can show them video of their transgressions.

So, after describing what we do, I reread your original request and would add these thoughts. I feel comfortable with our evening operations due to a variety of things, which all facilities might not have. First, we have a very solid hazardous gas distribution and monitoring/control system. In our older facility (pre-2003), we did not have as nice of a system and hazardous gas usage was limited to normal working hours. Secondly, we have a solid emergency response team on campus 24/7, and they are first responders for all of our emergencies. My comfort on night operations would depend on who would be first on the scene and how well they could handle the potential events the facility could have. Lastly, we have a solid culture of lab policies and enforcement of them which helps to provide a peer pressure to follow the rules. Many of the night rule violators we have caught have been turned in by other users. The video cameras help to back up anything they report. We obviously don't catch everyone, but we catch enough to keep control on evening operations.

I hope this information is helpful. Feel free to call if you want to chat more.

Dan

Dan Woodie
Lab Use Manager
Cornell NanoScale Facility
250 Duffield Hall
Ithaca, NY 14853-2700
(607)254-4891

Begin forwarded message:

From: John Shott <shott at stanford.edu>
Date: May 29, 2011 10:10:27 AM MDT
To: labnetwork at mtl.mit.edu
Subject: Re: [labnetwork] 24/7 operations

Matt:

I'm sure that there will be a lively and interesting discussion of this topic.  My guess is that we all probably have some form of "buddy rule" will differing approaches and degrees of success in enforcing it.

Let me get the ball rolling by telling your what we do in the Stanford Nanofabrication Facility.

Yes, we are open 24/7 but are only staffed 1+ shifts .... approximately 6 a.m. to 6 p.m. M-F.  We don't limit tools or processes.

Because of our physical configuration, the sub-fab where pumps, chillers, scrubbers, etc live are off limits and not accessible to lab members.  Same for the gas bunkers and main chemical storage areas so we at least don't have to worry about folks in those areas trying to "adjust" things themselves.

Our card key system is managed as a part of the university-wide system so we don't have real-time access to who enters the lab.  Folks that tailgate when entering the lab or fail to swipe the card reader when they leave would make that information virtually useless in terms of actually knowing who is in the lab at any point in time.

We have a rotating cell phone carried by several members of the senior staff that is for emergency calls for anyone in the lab and is also the first number that the central-campus facility monitoring system calls if sensors relevant to our operation go into alarm.  Things like toxic gas sensors automatically sound the evacuation alarms in the lab and, if gas sensors in breathing air go into alarm, call the Fire Department as well.  Being in earthquake country, we also have a seismic sensor that shuts off all gases and sound the alarm in the event of an earthquake.

Although we don't use this particular feature, since I know that you are running Coral, that gives you the ability to create and use either a lab-wide role or an equipment-specific role of "restricted" that would allow you to define policies that would prevent people from reserving or enabling equipment outside the period when staff was available. That allows you to let someone get more experience with staff resources available before you "turn them loose" with unrestricted access.  Of course, the downside is that you and your staff have to manage who is "restricted" and then determine based on time or more rigorous means when someone is allowed to work at any time.  Also, while you can limit when people enable a tool to a certain time window, it's very hard to control when they quit using it.

Are there times in our facility when people violate the buddy rule?  Probably ...  Of course, depending where people are in the lab, just having a second person in the lab isn't a guarantee that they'd be in a position to help in the event of a problem.

I'll be interested to read other offerings on this topic.

Thanks,

John

Begin forwarded message:

From: "Matthieu Nannini, Dr." <matthieu.nannini at mcgill.ca>
Date: May 27, 2011 1:47:54 PM MDT
To: "labnetwork at mtl.mit.edu" <labnetwork at mtl.mit.edu>
Subject: [labnetwork] 24/7 operations

Dear lab managers,

We are in the process of planning a transition to 24/7 operations for our fab. I would appreciate if some of you could share their experience and give advice on must-do's and must-not-do's regarding 24/7 operations. How do you manage staff ? any restriction in processes ? How do you re-inforce the buddy system ? Buddy system at all times or only for specific processes ? Any specific training for users willing to work overnight ?

Thanks in advance for your valuable insights.

-----------------------------------
Matthieu Nannini
McGill Nanotools Microfab
Manager
t: 514 398 3310
c: 514 758 3311
f: 514 398 8434
http://miam2.physics.mcgill.ca/
------------------------------------

_______________________________________________
labnetwork mailing list
labnetwork at mtl.mit.edu
https://www-mtl.mit.edu/mailman/listinfo.cgi/labnetwork