X-Virus-Scanned: clean according to Sophos on Logan.com Return-Path: Sender: To: lml@lancaironline.net Date: Mon, 03 Mar 2014 07:50:06 -0500 Message-ID: X-Original-Return-Path: Received: from mail-pd0-f173.google.com ([209.85.192.173] verified) by logan.com (CommuniGate Pro SMTP 6.0.8) with ESMTPS id 6754778 for lml@lancaironline.net; Fri, 28 Feb 2014 20:02:12 -0500 Received-SPF: pass receiver=logan.com; client-ip=209.85.192.173; envelope-from=frederickemoreno@gmail.com Received: by mail-pd0-f173.google.com with SMTP id z10so1419781pdj.4 for ; Fri, 28 Feb 2014 17:01:38 -0800 (PST) X-Received: by 10.66.164.104 with SMTP id yp8mr6706176pab.25.1393635698231; Fri, 28 Feb 2014 17:01:38 -0800 (PST) X-Original-Return-Path: Received: from User-PC (CPE-124-178-130-56.lns3.pie.bigpond.net.au. [124.178.130.56]) by mx.google.com with ESMTPSA id yg4sm24863306pab.19.2014.02.28.17.01.35 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-SHA bits=128/128); Fri, 28 Feb 2014 17:01:37 -0800 (PST) MIME-Version: 1.0 X-Original-Message-Id: <53113169.00018D.09660@USER-PC> X-Original-Date: Sat, 1 Mar 2014 09:01:29 +0800 Content-Type: Multipart/related; charset="iso-8859-1"; type="multipart/alternative"; boundary="------------Boundary-00=_H6GQQADPJDU1VA400000" X-Mailer: IncrediMail (6605288) From: "frederickemoreno@gmail.com" X-FID: FLAVOR00-NONE-0000-0000-000000000000 X-Priority: 3 X-Original-To: "Lancair Mail (lml@lancaironline.net)" Subject: Canopy Latch - The Checklist Fallacy --------------Boundary-00=_H6GQQADPJDU1VA400000 Content-Type: Multipart/Alternative; boundary="------------Boundary-00=_H6GQLLHPJDU1VA400000" --------------Boundary-00=_H6GQLLHPJDU1VA400000 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable "Not to put too fine a point on it but, including a pre-takeoff checklist item 'canopy latched??' should cure this issue once and for all." [Emphas= is added]=0D =0D Let's put this single pilot check list falsehood [emphasis added] to bed once and for all, and do it based on tests with real world data collected over many years. =0D =0D Time for some real world data regarding human performance. I learned thi= s from extensive work on production line quality control in high tech manufacturing. If you take a person in a good environment (temperature, lighting, sound) and that person does a task for which they are trained, experienced, and have the right tools such as assembling a complex piece = of machinery, on average (40 hour work week, Monday through Friday, all year round), the error rate in a simple task such as installing a screw to a required torque as you do so is....=0D =0D About 1%. That is, the measured error rate, over time for a trained and experienced person doing a simple task is about 1%. =0D =0D This has been verified in airline cockpit tests, production lines and man= y other environments, and seems to be a rough constant for human beings doi= ng repetitive tasks. For some people having a very good day, it may be a fraction of that rate. For others, harried, tired, end of shift, it may = be much more. But figure roughly once per hundred small tasks, an error occ= urs averaged over a long period under a variety of conditions. =0D =0D Same applies to a SINGLE pilot running a check list. Some days you will make NO errors on a 20 item check list. Other days, tired, harried, distracted, you will make a mistake. The most common mistake: you miss t= he checklist item. =0D =0D Bingo, you are tired, in a long line of traffic waiting to go, hot day, canopy open to stay cool, suddenly you are "cleared for take off without delay" and forget the check list item. For most items, no big deal, the airplane flies. For the canopy, maybe, just maybe, missing the item on y= our Legacy means you die. =0D =0D This is the reality of human beings and some of our aircraft, so the reco= rd clearly shows. =0D =0D So what to do? On the production line, standard practice for "six sigma" quality (one error per million operations) you do the following, standard operating procedure:=0D =0D 1) Do the task.=0D 2) Go back and check your work referencing the assembly task list. =0D 3) When you pass the assembly to the next guy in line, he/she checks your work, and then proceeds with the next stage of assembly. =0D 4) People rotate positions so you know you will check his work today, but= he will check your work tomorrow. =0D =0D Error rate is then 1% of 1% (the check) of 1% (checking the check) which yields one error per million. Desired result achieved. =0D =0D Now look at an airliner cockpit. =0D 1) Check list item called out=0D 2) Check list item repeated by 2nd person=0D 3) Check list item accomplished with call out=0D 4) Confirmed by first person calling out who does not move on until the correct challenge and response are obtained. =0D =0D Error rate: one per million. Or less.=0D =0D Now look at YOUR cockpit. =0D =0D 1) As you go down you check list, you get to item X, and miss it. Chance= of it happening is 1%=0D 2) You proceed to the next item, not knowing you have missed the item. =0D =0D Overall error rate: 1%. Repeat 20 times for your 20 check list items. =0D =0D Your error rate, compared to the target rate of one per million, IS AWFUL= =2E=20 Average single pilot alone operating as an individual in his cockpit with= out internal checking and observer operates at this level.=0D =0D Not every day. Not every flight. But on average, year in and year out, you will miss a check list item 1% of the time. =0D =0D And you could die as a result depending on the item missed and its consequences. =0D =0D NOT GOOD ENOUGH. =0D =0D So what is needed is a FAIL SAFE method of protecting you from the fact that you are a human being. If you miss a check list item and it could k= ill you, you need ADDITIONAL LAYERS OF DEFENSE to assure an adequate level of safety. =0D =0D A virtually perfect defense on a Legacy canopy is a secondary safety latc= h that automatically prevents the canopy popping up if the main latch is no= t secured (auto hood safety latch concept). John Smith's analysis showed t= hat conclusively. (Plug for John's work: he has worked safety issues in the off-shore oil and gas industry. Remember Piper Alpha. Much was learned from that and other accidents.)=0D =0D Otherwise you are working with layers of imperfect defenses: your own secondary check (push canopy up automatically, muscle memory), audible al= arm (different sensory input) warning light (easy to miss with sun behind you= ), big warning message across EFIS display (hard to miss when you look at ai= r speed indicator prior to rotation) or whatever rings your chimes. These provide vast improvements because they catch you when you miss. Most of t= he time. They add layers of defense, and reduce the probability of a fatal miss from parts per hundred to down to parts per million=0D =0D Yes, we should all be good boys and girls and always run our check list, every item, ideally with somebody challenging our work. In reality, we a= re human and the environment (outside and inside your body) sometimes interferes, and you make errors. Guaranteed. =0D =0D If it means your life, a checklist is NOT good enough if a missed item le= ads to a condition that is unrecoverable. DO NOT mislead yourself in thinkin= g that it is. History has proven that checklists without discipline and challenge are not good enough. Humans make mistakes. =0D =0D Fred Moreno=0D =0D PS=0D =0D Recommended for retractable landing gear planes:=0D =0D 1) Check three greens when gear extended - don't take your hand off the handle until you see them . That can be a powerful check by itself. =0D 2) Three greens on base. =0D 3) Three greens on final. =0D Overall error rate: one per million, until you are distracted. So when d= istracted internal alarm bells should go off. Then you can stay near one= per million, not one per hundred. --------------Boundary-00=_H6GQLLHPJDU1VA400000 Content-Type: Text/HTML; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
"Not to put too fine a point on it but, including a pre-takeoff chec= klist item 'canopy latched??' should cure this issue once and for all<= /U>." [Emphasis added]
 
Let's put this single pilot check list falsehood [emphas= is added] to bed once and for all, and do it based on tests with rea= l world data collected over many years.
 
Time for some real world data regarding human performance.  I l= earned this from extensive work on production line quality control in hig= h tech manufacturing.  If you take a person in a good environment (t= emperature, lighting, sound)  and that person does a task for which = they are trained, experienced, and have the right tools such as assemblin= g a complex piece of machinery, on average (40 hour work week, Monday thr= ough Friday, all year round), the error rate in a simple task such as ins= talling a screw to a required torque as you do so is....
 
About 1%.  That is, the measured error rate, over time f= or a trained and experienced person doing a simple task is about 1%. = ;
 
This has been verified in airline cockpit tests, production lines an= d many other environments, and seems to be a rough constant for = ;human beings doing repetitive tasks.  For some people having a very= good day, it may be a fraction of that rate.  For others, harried, = tired, end of shift, it may be much more.  But figure roughly once p= er hundred small tasks, an error occurs, averaged over a long period unde= r a variety of conditions.
 
Same applies to a SINGLE pilot running a check list.  Some days= you will make NO errors on a 20 item check list.  Other days, tired= , harried, distracted, you will make a mistake.  The most common mis= take: you miss the checklist item.
 
Bingo, you are tired, in a long line of traffic waiting to go, hot d= ay, canopy open to stay cool, suddenly you are "cleared for take off with= out delay" and forget the check list item.  For most items, no = big deal, the airplane flies.  For the canopy, maybe, just maybe, mi= ssing the item on your Legacy means you die. 
 
This is the reality of human beings and some of our aircraft, so the= record clearly shows.
 
So what to do?  On the production line, standard practice for "= six sigma" quality (one error per million operations) you do the followin= g, standard operating procedure:
 
1) Do the task.
2) Go back and check your work referencing the assembly task list. <= /DIV>
3) When you pass the assembly to the next guy in line, he/she checks= your work, and then proceeds with the next stage of assembly.
4) People rotate positions so you know you will check his work today= , but he will check your work tomorrow.
 
Error rate is then 1% of 1% (the check) of 1% (checking the check) w= hich yields one error per million.  Desired result achieved.
 
Now look at an airliner cockpit.
1) Check list item called out
2) Check list item repeated by 2nd person
3) Check list item accomplished with call out
4) Confirmed by first person calling out who does not move on until = the correct challenge and response are obtained.
 
Error rate: one per million.  Or less.
 
Now look at YOUR cockpit.
 
1) As you go down you check list, you get to item X, and miss it.&nb= sp; Chance of it happening is 1%
2) You proceed to the next item, not knowing you have missed th= e item.
 
Overall error rate: 1%.  Repeat 20 times for your 20 check list= items.
 
Your error rate, compared to the target rate of one per million, IS = AWFUL.  Average single pilot alone operating as an individual in his= cockpit without internal checking and observer operates at this level.
 
Not every day.  Not every flight.   But on average, y= ear in and year out, you will miss a check list item 1% of the time.
 
And you could die as a result depending on the item missed and its c= onsequences.
 
NOT GOOD ENOUGH.
 
So what is needed is a  FAIL SAFE method of protecting you from= the fact that you are a human being.  If you miss a check list item= and it could kill you, you need ADDITIONAL LAYERS OF DEFENSE to assure a= n adequate level of safety. 
 
A virtually perfect defense on a Legacy canopy is a secondary s= afety latch that automatically prevents the canopy popping up if the main= latch is not secured (auto hood safety latch concept).  J= ohn Smith's analysis showed that conclusively.  (Plug for John's wor= k: he has worked safety issues in the off-shore oil and gas industry.&nbs= p; Remember Piper Alpha.  Much was learned from that and other accid= ents.)
 
Otherwise you are working with layers of imperfect defenses: = your own secondary check (push canopy up automatically, muscle memory), a= udible alarm (different sensory input) warning light (easy to miss with s= un behind you), big warning message across EFIS display (hard to miss whe= n you look at air speed indicator prior to rotation) or whatever rings yo= ur chimes. These provide vast improvements because they catch you when yo= u miss. Most of the time. They add layers of defense, and reduce the prob= ability  of a fatal miss from parts per hundred to down to parts per= million
 
Yes, we should all be good boys and girls and always run our check l= ist, every item, ideally with somebody challenging our work.  In rea= lity, we are human and the environment (outside and inside your body) som= etimes interferes, and you make errors.   Guaranteed.
 
If it means your life, a checklist is NOT good enough if a mi= ssed item leads to a condition that is unrecoverable.  DO NOT mislea= d yourself in thinking that it is. History has proven that chec= klists without discipline and challenge are not good enough.  H= umans make mistakes.
 
Fred Moreno
 
PS
 
Recommended for retractable landing gear planes:
 
1) Check three greens when gear extended - don't take your hand off = the handle until you see them . That can be a powerful check by itself.&n= bsp;
2) Three greens on base.
3) Three greens on final.
Overall error rate: one per million, until you are distracted. = So when distracted internal alarm bells should go off.  Then you ca= n stay near one per million, not one per hundred.
 
--------------Boundary-00=_H6GQLLHPJDU1VA400000-- --------------Boundary-00=_H6GQQADPJDU1VA400000 Content-Type: image/png; name="SENDER_EMAILfrederickemoreno@gmail@@com.png" Content-Transfer-Encoding: base64 Content-ID: iVBORw0KGgoAAAANSUhEUgAAADIAAAAyCAYAAAAeP4ixAAAAAXNSR0IArs4c6QAAAARnQU1BAACx jwv8YQUAAAAgY0hSTQAAeiYAAICEAAD6AAAAgOgAAHUwAADqYAAAOpgAABdwnLpRPAAAHJtJREFU aEM9mneYnNWV5j8JAzOzZoiSkIQiCgRhwCIahxkMGHsAD8H2DM4TzBqbpNQ555xTdc5V1dXVFbq6 q7q7qjpU56BWKwcMO7v/7J/7eGfWaxDv/m6JWem5T32p6zvvOe8557331ibry39jM6u65eZbreub /2xt/vNN1vVbPreufy7r5ltusrbdfpt161/ealmff2Fdt2T9xS2brb+4+S+tzz/7zPq/n/2Z8Zml 619YX7n1FktffGH9+2ey9Kf/sL64+Sbrf/+v/2P9ib/79z/+h/U///hHa9OfrlvXb/5K/Ls+++Iz i7+wPt/0Beebrev63Lpp001Y9Dn/LWuzNltf2czzX3zFunWzrD/dJGvTdcvatEnc3WRt3rzJev+X P9n0nxgsx8i8POOLckUX5ZtckH96Sb6pRT6XFZxZ0lhsSWeuXtHauQtaO39JZ65c0eVr13T100+0 cfmKls5f0PIF7l25pI0Ll7R09rzWL17U8vnz8k/NqHdkWq3uUTX0DKu226W6TobNrfJmp8paBlTa 3Kvi5j4V2/pU0tqp4qY+ldr6VdTcrwrulbd2q7KlVzU8U8a9ipYerneruJXn23oMKsvq9E2p0z8t V2BajtCM7GOz8kzMyB2dlDsS02BkVr7wrCKLS5qeXVN0bkVTiwuaWTut81cua2HltGZXVhXjM7a+ obnT65peXtNpQC+eOStPeFrd/jE19Y2ossOl8na7StswusmhEpsd47uVb+tRcUO/Suo7VFTfrYKG LuXXtyuvkXMbIBklXCtqbFVhQzt/16Hixk4V8rdlTdxr7ZXV4Y6oYyiqLu+kenxR9fsicgSm5Bid AlRUA6OTGhybxqApucOTGp7keDKGp2c1EVtUZBZgc3OKza9oduk0IDYUW1pTeJ57M8tyjQPEO66a Lh8AXCrBy4VN/Squ71d+XZey67pViNG5td3KretQXm2bimq7lFfTxjHG1wOqrkUFjbY4wLIanq0H RF2bCurbVNHYoxJAWS0Dk2oeCsvmDgJoTG1DE2r3jKnXNwGwsPqGI3IOTwIuqsHQ9JcjJh9RGwnH NDw1qZHJeY3MLCg8PaPx2KxC0wvyRac0Ep0HfFStAyOq6hxQSQtA8Hwmxmfg+Zw6W9zwnNoO5dZ0 KKe6VVkAyOQ4vYrz+hYVc15YzTP1jcqpuXE/m+fza4kYESwGTD5RstpdE2p0jcs2EFQLw+YyYEbU AbAueN3pHZMLj/b5o7IHwrIHiVJwQl48PRiKaogoeQEVmJgjz8LyjM3IF5mTHzoaEC4i2mEfVgVA ClvsGGS83U4k8Gx1yw3Dq9sxvlkZla3KqGpRWkUroLjHyK7sjAPM5Zn0SptyuJdbx7O1LTihhejw LIOIRNTcH1ITXmt2htTmDKp9cAQwY+omMl1DQfV7Q+oJTMhJVJwjjNCknGNQDVDOkUn1D4+rPxDh b4Z5Jsx1cs5cD0yqhe8q7XSqsNmugro+xg0KZdV0KqOsRakVAClvA0CT0jnOKGlSRkWLUirrlQmw ZJ7J5TwTEBmMdK5lVzUqu8YGuAYi0wLQZlm2gbE4rZrdgOgPqN7lUx/HJiIdg2H4HZJ9dFzO0aAC RMEzTv4whojCIKD6/RPqhI4drgDOCDL8anUEcM6oqrs8qmgdVBGVJq+xP258fjV50dCEt9uVBX3S q5oxGkClLUovtSmllHNGJuepFRhfYuNes1J5Jq2yWTnlPAegTI4LGRnldQACSKcfr7lCeC6oVoaJ TOtASM2DATw9qbHZWQ2ORjRkKEMF8plKNjqjAQqCfThMFEbV7BpVVa+XPPCoklHR5qFkDlJaXfES Wtxoh8d9yiShDZjc6s54DmRUQCmikFbWzGhUallrHEQSxicSmcT4Z71SyhuUznlKca3SixuVxnkG IDMryRuomG2A9FC1Wp2janQMq9k+Aig8Cr1auOaemFIgitHjM+RELF7BekejUC6sTk9QnW4o6Aqq sX9Ypa3OeF8oaOxTDiOXipRXg/drewh/pwrIi0yMzyRxs6pIXs5TysiVsjaMx9vFRAPDkzA4zYDB 4OQijgsaAdCg5BI+S82wKauwUenlNUoHfB7XMqubSHY3Fcs5rBbHqJqcIwyv6uw+NQOs1TWs0dgN EA5DI5K+ExBtbqhj9/OsXw12ryrbB2lQ9IhmB57vp5TCf1NdqvA8Xk4ynjeJW1ZPIsPzWkY1A/6n lOF1Y2gRn4VNSips0ImiRiUU1imhpAZAtUourFZGYT336/VRRo6OpWfqWGo6gOuJbB2RaZDV6x/H wxNqGzBRofs6/Krr9VPFhinBITnJgS7fKDlAtOB+K2Abez2q7/aoptuvyrZBldtcdFkH0ehXXl1v nEKmhGaRpKcq4Hc5noQ+BlQ6HDf5kQ4t0iuJRkkLxjYoEePTSwFS0qC0IkBgfFoRAAtqAWDA1BDZ Jp1IT1NiRpZOpWUrDaBppQDBQVavj3JLjrQzGohMLTRpdvjw/hhJTO4QnaY+PN9HIejzqLrTrRpG VbtDRa12lZDIRVSk3PoevAxlSOYUACSW30jYU+VwvJDqA2VMQicSAVOdcqoAA62yi0xekNAmKgUA Mcbj+WQApRTU60ReuRKKqnUyr4brRTqenqrjmTk8W40TAF98A4xlgxrNhlr9XjU6faq3e0j8YUqw oZtf9V1Qp9ut0g53vAJVxhPYTseFQpTSfJpWbhVJDP9TMS61whjcCEWIApxPzq/HuHolVHBuruH1 1PJWngFA8Q3Dk+KRqFMyYJKIRjL0SizH+JLa+DMJRCOjtEzJWbn6KDVViTmZSuY8yYDhuxPJJ6u6 z62Gfrzu8JIr0AvuNxtwAKtH4FVAGeN5I+TykAK5ca+TpFVdJK+p+dCjDDpwLZGETcSIlDyAYFBa SbNOwfWTBVDFeL20Fs8Djpcnc/94fgWer8EYnsdgU5WSi2qUSIKnU62SigBiaJXXoMziEiXmpSoj M1OnsjOVlAewwiolme/g06rpHFJDr1v1PQzqfhPKtBZdVNtP8ra4EW09aJqeeCfOhtMZcD4d2qTQ mFJ4mTEsAQokFjN4cXJ+rU4wEhinGMfjhtTh8Wo8brgPCKKUkFcLwCqMq+IY+sQBV+OIcv6O5/N4 vqCSzxocVgWQYqVnZemHr74JvdJ0oqA8/rcnCytxQKWsyg543zFA0jpUhffLOvoB4kLkIbGJgimj WXWd8ZKZTgdOJYyJJG5qaQMUsVGF8KRpZkgHU1WMgadyMSzf8LocjtfoRHa1jnN8HCOP51VyrVIn uXYMbyZlVenDHM75m+P5lTqejYE5JXxHjY7nVsVp+t7xj/TTn/1U//TOP+mff/MbfZSegTOqiVhl HKyJpJVjM5IZgdbchYrsIgp9qkL7F3NcGlenUKIc7hc3K8vUckpkYhkhN00KYGlUpDST2NR0k5yJ GHc8uzRu+Ae5FRiFkZnF+ii3VB/mFegD7r2fVa5jjPczy3Qio4z7pTqWWcJxiT7g2rEsc61Ix3Lz oDHVqqZGhUVFSs0tUlJOtj7MzABgBYluKlsFFAZIbly0wfu6VtRkO5rfTFh64lrfaJgMpEEGXkk1 9T6erOQAiZtFfmShRjORGBk8l1ZJwvLFCfD1RDZezypE1tB3+pxx2jb1DKnH5VZqfpEOHnpAu/cd 1radO/TEt7+j9xMSoVCx3s0u0XtUpPfS8/F6vvLpEcX19SqpracHlcaBnIRW6XkVSBVD6yoqXwWO pvxm1GJIDcILzZ/DRKbA1qmiuqa4ssxG82dV1yuLZP6IP04haVMr+FLUaAE6Jy7YalvRT6bhNcUb XlN3n77/5o/0taPPKq+sRnsfeEbbDn5Nhw49qILqWhXhgCanU/0eryZnV/V+YjYUK9XvU9L03Le/ rfsP7AfoQR157DE99MhRPfzIEe176GH96Gdv6/eJSfrX37+rjGKiUE5E6O5ZgMgiV618GlQ2Xs1D 1xt5XNCAvm9ifoDczjHXUJkJhC8hv0yn8oqQFkQDAPmAz6tvij+TZSQ5arWqplKv/cOPdejhZ5Re VKJdDz+nv/vpz/Xqj3+h1JJiPFiOFCmFhtUUjnIUQDWTqHrKvEvtjG6XR7VNDUj1Sn2YkqoXX/uB nv/eS/rmd/9GL73+in773rt65/gx7T58WAceflAHDzyovQcP6sW/e02WQWQEWA5tPhu+51ahZaoa VBSPEsldUk4ilpKcxTqekYu4q+HlqFDuFRHFPEYueZRfW6evP/O3euJbz+uNN35CFCu18+AD+uWv AZGXT+LnKZPvSqOMJhQWwu1iHFREJYMyNLq0AmiTk6cP0tJ08lSSfv67d3HAj/S91/9er7z5hl7+ wct69tlv6Oknn4t/d3ZVLbZW8/5GFdLxLaMqDddSSeC0ijrChhgjJ9LplmnQKAdwqVSTEyn5ei8l XQk0pXwMLzQRMZGpo4cUliAOq4liLWEu528IfVGhUgsKlUTCJgMiKTdHidlZfGYrISdHaRidmp+v hPR0vYPRb735lr7znef19ccfYzypR7/2NT3wyKPad/Cwth98VNv3PqIdDxxVVV0tasKm0vZ2FHaH KptbqbjMEA8ePqI0EqzMhgI13ZKqk0ouJFP306jRaQZkCfWe44/QOMfSMtVFAle3sBAAiKzSYuWQ 5FnQoZColPGi7BKqUUKC3v7Ht/Xd51/USy+9pBdfeJHxfX3zb5/X09/4ph5+9Clt27NPt9+7S7fd c5/u2LJHt23Zq9u27tZd2/fqju37deeOfRzfp0efeFZtvd2q6upSS2+vbL19svX1oQ0dzH2gpdMh 69U339ThI0f02NPP6NDRozryyCPad+iAnnz6KY0GQ6pqomoZ71Id8muYDxTk61gOpTMlQ2+89ZZ+ 8P2/16s/fF0/eOWH+tYrr+uhx1/QXfcf1T2Hn9Y9+49q64Gj2rbvCW3Z93XtPPAUxj2Awft12917 9NW7d+qrd2zn+F7dcfcu3X7PDoDt0R1bt+id376LwrCrGaNbMbh7wKWeIQ8a0IuYHWE+NMI0ekTu UabfIwFZqXTIhPgohMfVTEVtevmN17V7z17tv3+vdh/cr/0HD3B+v+6552794pe/0OVP/01H/+YF 7dj7JAYe1a5DzxL6J7Rt7+Px860GwP1PavuBb2jL/ie0nfPbdh0CyDd17/5ndNud9wFgm/7qr+/S V+/cBpBtunvHLn144qQ6XYPqdnvl9Pnk8A4zBxrV0Aiz01BQ3vC4vNGohs3CBosewalpBadjLHzM yDqWXaSE3AIoVQx/SymxlXHJcCq3hCQtowGSUJS5t3/9K4zdpx2H9mobJXLXgft13759unfnLu3i +ms/eVs//PnvtAPj733o29r74HPa+eC3tPOh72rrniN4/17dfudWjN6q2+/aptvv3gKdtmnL9p1q aGlBfbvUNWhUt1dej1+DftYKhkPyj45pJILR01MsFs4pOhNTmDW2qfl5TbDkFJxf0+jcsqzjdNRT ADlVUIB2MQAotVAnKR+pUGgUJqOoXKlFVXFNczw1V/v279P2+w5p96GHiNg+orVHO/fs0pa992nb 7j26fQsG37Odz+36L1t3wvOd+uttW3XHtm26c9sO+L9T9+7aoVdfe1W1bR2ydXertrVF3VCoa8jL GtuQXD4Pk7kRDYXG5GexcAzPj8+x5DS3qPDCgsYBM758VtGVDUXXNmSdzKeq0DGN8af4PEElSabD nsop5JoByT2od4pxIjtHP/7Vr/X0Cy8jp1O096Gn4PtOvfDyy/oGHfrAkcPa88BBAB6ivh/WveTa rt17icge3bN7t7YSubsAuv/BB7UfJ+zk836a3RFy9MHHHtfXn35Sz33nu3r5R2/qrZ/+o/7lgw90 /FSi8osLqZLVsrW1sUTlBmCIRZCoxibnNMbK5yQLg1YS/eFU/o3abj6TzShEI+WWxfVSApFJNkqT Wv+7hCS98vbP9L03f6zkVOYEBcX0D7orzS+rpBTNU6nHn3hKO+7bxRJOtQrL6uhHdVSzWiqcDRHa RPGgcba1smbbovrmdiZqHWogInVUoWYSugNDu70edfn9co1BrakphVjJjLAIMs9y7MLZc1o+e0FL Zza0yNrymWtXtXLpMsmO51PLK6FOGXOAEkowUoQ5QXIpEhp1mQioNDpxEhrpnY9O6l9++1/1D7/6 Dc0ri/yhmZWXIxXKVVBVSYNCgtCgqvBcbUeH6np64kY2UPMbMb66owfDMZpr1T18dnSqsauXhcEB ZqmD6nFTlXzDGgqOaSISZr15XrPzC4ouM5aWNAGlwkurnLO+zHrzAsDmz5xm0fyMrHq+uIH6XE5f qOFlDZ1dvKxTZYi1ooZGRgPHLJ4hKVKycpSQnK53k1OZo9SikGlMDS0s+7Sywt6h6u4e9fT2yx+k yhh+j4+zAhnWKJXGPxmRJ8qg8viioXgF8mC0LzQhf2RCwckpRaenWTOe1mRsXjOLi5pfZ4F89TQL 42taPrOuqbU1InJes+wILLMDsLB2TitmF4AdAKuPKuEa8cs9TKnzszhHkvX7XHIOetTuppIM4Cm8 ZapJa18vM8de1Gyv2gccrBVDAfcQC3luVG6/Opx2yiYe/bJUBqfh8cw0a2MzCsViilJlwhg4uUzV WV7RDF6eWlrU9NoqxrKiv7aiOY7nV/H28rKWDIiVFc1vbOD5Da2cO6vTFy/ozKVLWt04p1WoZcbK hYuyBkd4MQ3FNRyQg2H3+NSLMu2henQOuDU4OCi7j7rOPT8eHp2a0Oh0RBPT46zMszqPtwPRCbkw fig4ytrvGMtHVJqJiELU+TFGBHpE5mPaOLeBERd06dplXbpyUWsXz2sRzi+cPqNljJ9fWdYs3jdb FrHlVc3EtyvWtLhxVmuMFYw2NFo+R56cB8gZc3xWywC13ENuIgEQlkRdAYD4vRrA03aakd3nl8M/ TCkMxJuSPxLRKM0nMjeryfhYVAiPh2dmNc716FxMc/A4gpdj8DvGcZR7E5TNaSLx6f/4RP/tv/9B 1/5wVR9/+rEuXLmqs9euaAMDjcdnoU7MgFhd0TRbE5NLy5qZZ18GkIvrZzTD9QWALmL44nkAnD6t BSg3y99anhClzFCBBPNOTNM5ZzU0OasR+BoIR+iedM6pOQVm5zCaYXg8N08iLmhqYV7jNKroHCBW MHYB2phNIJrVNPemARuh4pjryxvruvCHa+x0Xdbljy/pk3/7lF2wy7oKmLUL5/E4m0Srxvil+GZS lD4RgYKTOCUKoDkiMwkdDeXm10lykn6KayZ6S+trRIR67GVrwDeFURg6TOMJzCxqgjFO0oVmjQSI QZE5kpbP6cm4cbOUw7Apies0JnO8gNFL81pc46Vw3QAaibF3wt9GOb546Zyu4v3LH1/THz65oosf X9FZuL4EwDnjWXrBOAaPY/gEThshp0Js+40bp5lNIyI6gYMmY3OanJ/BmUR7bim+oTTPDpkVYm13 lH0+M8Z5MERijgMigHEBvsjsIZoNG5/ROAD2jEc0zt9MADBqqDPLwjb3wmgfY/wsXpzDGNN9g/SA cYAswPdZAG5cgOuMs+wvLkKR6WWcxTtNhP0Y7qODe1hnHmJfZShidBXSJIKuYpcsyCZSwFQ/GBCC rjc2klibJk/9sMSKzvJCviQ8Mx83bIIoGL772Xka5Q/8ZhGb/HD6GMMkcyCkALliktkTnmAni7xC 2A2Hw1+CmydP8B7GBXnJRGw6HpH5dUOJM3B9XZMLyA0cEcKoQISdMOb2brq1c4QNJfRVr2+cd46x scRu2YgpJGPyUWgc6K7BYJgqG2Y9gOfZcBow52gxKwhNJgAyRqhCeG+cResxttGG8aabPUS73+yV UIK9PtaBR3npKCAmNUp/8PDyTpSqgyo3SO8YoVdEMHCaaEb5zgg0NNsSMZJ1jkjNUgCi83NEioVx AAxgoAsn2YeDOGRcvXx/h38EAFxDNPYZB7KR6g6yJ8NGk4O1aAeVcYD7DkA6AmPIlbAG2auxYiRZ lPDOUCLDgDLJ68Xjdjw9MDyCIg2oze6m+zKBoae0Uo676cD2YZ9ah+ghHLsAEaTZBaaihDusMBQY IxIm9FGoanJqjIrm57qbKA5QYOx4eoC5hMMTUjd7LN3eUXpSgAV1v7pwTCdzji6qaR8bTQ4i1Yst PYCzA8hOZLzjE2zSsm/DppM3tiLL8MuU1FGzqcnwoG+MhO5xD7NCz8ojq5BNLKvamBXael2MAdWx CtnUP8gS6yBNk/5DmXZPjNFMfRoNh4gE/cNUPfqLiewooQ8B0ETSw8s90GGAaDrYq+wbCrBY7pNt yK+OgSHZWHPuYguvi+j0AWQAhxpHGQoNoQJ8RNLMR4I4fnxxVSMLGwoyrKExKEL5dU8QOnjp8OIV ekgvoq0X77SwldbIWnCbY4AZm1OtPQPImH6WfQDGFLMd2rUDut0zBPAhDKMP8fIhHOIdG1fQUJD8 8REJz1gQL0InqNHrGUYRDMWj0I7hnXxHH/2qz0OEoa+X2alxsNkxDhn5TrkNL61QuVYVottPsocf Q6KMLPGjBIZlOrMbhK6JKD8YIAd4acgkMxQbZJe2dYBVerzfxLSzwe5SY4+TtWI7gq8LqQKofjMc CEQGCrahz44jfHgVYEicPqLVjuwx0erDQd3ImTaMbQdwr5l7sHXhIkf89KxAXM7E6B30JLq86UUT FJ0Ro34x/vQFOvqlizpH6T59+ZIuXL6sORriAp3eipB8w8gIM3kZMclJpTFTR1M+p6jRMcqkne7e aXapnFCM1cJGJvz16C6b00XOuKGEV932AehGhIhah8OJvHGyz8JPNaBlk2OIBQJnXOG2s7XXDp2c fKeb6AybAkOVm/6yq8/QU2Ko2nWapJEmM2vrlOr1+M9E1pE3BsB5xiWUwZVPrmoVMCuUc8uP6vRR t300u0E42M/GpzvEFjM9w+RNCFBeknjQVBhKnpuk9rO3aPbaB0PUelMa4b8Pjw6RgB6+awCV0E+y migYYdnNZ7+X6auJNE4zRSBkmihdfGaBvoPxRnacRj+dvXJOG9cu6TK/fbn08WWdwWjzm5YNmuk6 v3W5wPWLzEHMb2M2UAXrF4kKkbKM8Wa2ZSpJLxWki3LXzy5WB5To4uUOjns47mb7rR0O20lAhymR I2yWwmM/uRVg7hCajEIDKtTiLJRYiNPDqNyIkRzxLo3DaGym6Zl9yaABMj+LCgYIcn3t7FmdQ8We Ach5BOWZK3ifCJjrCxtnNEd0ZgF85qr5Uc9FrZ/f4Ec+5+IazYhIqx8DW6gY7VQOsz9oY9eqacAH fXwk+hC7V+bTG9/FaiMhOwHVR0HoMUKTpDXzCyPVIwbEAosDgJgxEyIjHinrAcpwIGL6Bk2W37OY GV8ACgdNZTP9huqzQJ9ZOoeSvUBULhKRixs6C4BVhGEMWT9NkpuJ1RRabNrIf+hu5ijzSH/zgx6j mOO/EGqmTzSx5dZOUtf3UwLNJzxvpvw2AsokZYeXn3VAkS4qWTdg+qk6duYwLsqom8oUiJfY6XjH jtAzxoxgxONxyUPCmknVgKlYRN8Dbf1ExyjnGXI0ukpio2znT69olegsm/xYx2CuzWL4NFGbAEhc DSMSjRqeW2WixXwlxv3//3stc1DDzyyquwbi2202tqhNVenG2F5PgGbEpAuDDQgzL+kxDYvhJJL9 nDup8x4khGl4Huq80V6jUM00Rx+fQ+SPySdn8MYvKMza1Bh0M+XVADXaaQrdZXJlDuMnAWMia+Yx Rg0bpW1oagSlATNlSrFZEiIa/wni/wHgNWODea0WGQAAAABJRU5ErkJggg== --------------Boundary-00=_H6GQQADPJDU1VA400000--