 Hi everyone, this is Doug Miller and I want to welcome everyone to our fall 2 Version of geography 596 a the capstone proposal process We have three evenings of talks this week tomorrow tonight tomorrow night and Thursday night, I guess this is Tuesday At any rate, I want to welcome you all We have a kind of a sparse crowd. Hopefully we'll have some more folks joining us tonight And I know all of our students have been busy working on their proposals with their advisors during the semester and I'm looking forward to To hearing what everybody's been busy working on I'll point out a couple of things to you We we ask that all in attendance not just a student's taking the course Fill out the evaluation form. It's in the upper Right-hand pod under notes. There's a link there for you to fill out the peer review evaluation form and again It's it's a bit of a misnomer. We ask that faculty staff students casual attendees take a couple minutes to fill out the Fill out the evaluation that goes directly to the student and provides them some some feedback Additionally, we have a question-and-answer session at the end of each talk Our presenters will have 25 minutes to present their work and then there'll be five minutes for interaction Speakers I will send you a private chat message When ten minutes remain five minutes remain and then one minute remains and She'll check that you'll get some sense of where we are with timing Our speakers tonight are John Srucci, Josh Dewey's Chris Dunn and Matthew White and our first speaker is John Srucci. John began this semester as a gainfully employed person and then in the middle of the semester he retired from his first career And he's now jumping into his capstone work over the next few months And it was great to get him started. Justine has has done the bulk of the advising with John but it's a really interesting project and the MCDA work is is actually something that that We should be doing more of in terms of trying to quantify decision approaches. So John I'll let you get started when you're ready and Take it away Okay, thank you very much Doug. Yes, this is John Srucci. I have the coveted first-time slot here tonight So I'll get things rolling The subject of my capstone project is retrospective GIS based multi-criteria decision analysis fit of a mouthful I'll try to explain what that is and describe the objectives of my project I've selected as a case study topic the sighting of waste transfer stations in California And my advisors are Justine Blandford and Doug Miller There we go. Okay. So this is an overview of what I'll cover I'll go over a background on multi-criteria decision analysis theory MCDA Talk about its applications with GIS based MCDA and then get into the objectives of the work that I intend to do Taking a retrospective view on GIS based MCDA I'll cover the selection of my case study the methodology that I'll use Expected outcomes and the timeline for my capstone project So I think we all Appreciate that we're making decisions every day and most of those decisions entail considering multiple criteria The decisions are usually intuitive The process is simple the criteria are implicit and more often than not there's just one person making the decision When we talk about MCDA we're really describing a collection of formal processes that take explicit account of multiple criteria and This these methods are especially applied when the decisions are complicated They involve perhaps conflicting Objectives often there's multiple stakeholders involved and ultimately the results have a high impact of course many decisions are spatial in nature and Perhaps many of you have been involved with using a GIS to provide spatial decision support So it's quite logical to marry these disciplines of GIS and MCDA and that Intersection is in fact an expanding niche field. I've shown a plot there in the lower left corner That depicts the peer reviewed articles on GIS based MCDA. This was as of 2006 There were about 300 articles that number is now upwards of five to six hundred This is a graphic showing the general MCDA process Of course, we start off with identification of the problem And then I go through a problem structuring phase This is the point at which we identify the criteria that will be involved in the decision process We explicitly identify alternatives. We characterize the uncertainties with our criteria and other constraints We identify the stakeholders who will influence the decision and other environmental system factors and constraints These all feed into then a model building step, which is really the meat of the MCDA process and in model building Variety of different kinds of decision rules are applied to develop a criteria preference and aggregate That criteria data into something that tells us the rank or value of the alternatives And I'm going to take a little bit of time to deep drill decision rules The model then has an outcome that provides information to the stakeholders They have to process that information and synthesize that into an action plan So getting into the decision rule models now I've taken a rather simplistic view of categorize them into three different buckets and this is simplistic. There are really literally dozens of different MCDA methodologies that have been developed over the last couple of decades and they don't all neatly fit into these Categories, but more or less this is a way of looking at at them And I think it will serve us well as a basic starting point for MCDA The first type of MCDA decision rule model is value measurement. This is perhaps the most logical intuitive concept around decision-making For a given alternative The criteria are assigned a partial value that may be a direct measurement of something that can be Determined objectively or it may actually be a stakeholders subjective value And for all those the criteria the partial value associated with an alternative then is aggregated in some way To derive a value for each alternative and then once the alternatives have been valued They can be ranked and compared so this follows a rather linear logic and And it you know, it's easily handled by a lot of software tools Certainly ArcGIS has a lot of tools in the tool set that apply to this lends itself to to raster algebra and overlay techniques Unfortunately the critics of value measurement point out that although it gives you a neat answer It doesn't always accurately represent really the subjective valuation of stakeholders So other methods have been developed and proposed second category is reference point models and in a reference point model the there for for every criterion a reference level is Identified and again, this might be an objective measurement or more than likely it's a value a subjective value of the stakeholders and Alternatives have to satisfy that reference level So starting with the perceived highest ranked criterion Alternatives that don't meet the reference point are eliminated and we march down through the other criteria and do the same So it's a heuristic approach. It's purportedly more the like the way that people actually think about making decisions because it has a number of binary Switches in it a Boolean overlay is very apical here on the downside You may not get one alternative at the end of this you may have more than one or you may have none So to really use it thoroughly It has to be iterative You make a pass through this and then you go back and adjust your reference levels And as such it's a really a good screening Technique it may not be appropriate always for rigorous MCDA the final category. I'll describe is outranking Outranking results in a a rank set of alternatives and it's a rather rigorous process where alternatives are compared pair-wise for each discrete criterion and for a given criterion Alternative is either determined to be better in different or worse than another alternative and again This is a stakeholders value judgment So the stakeholder doesn't have to assign a value He or she just needs to make this determination And then we march through all the criterion for that pair and then repeat this for all the pair combinations of alternatives So it's a way of eliciting Stakeholder valuation. This is what your eye doctor does when When you get the lenses to look through and is it lens one or lens two which is clear lens three or lens four You're not assigning a value to the clarity of the picture but by comparing two things you essentially are Are doing that and when you take the results of all these pair-wise comparisons then that can be aggregated in a number of different ways To develop a ranking of alternatives So it's highly interactive on the downside that involves a lot of involvement from the stakeholders It's very labor intensive and computationally intensive. So it may not always be an option for some decisions I'm going to take a look at two specific examples of GIS based MCDAs In these next two slides the first example I'm pulling up is a some work that was done this year or published this year by Lawrence Berkeley on The land suitability for a GAVA crop used for bio and energy feedstock and they applied a value measurement Method actually specifically the analytical hierarchy process method Which some consider to be a category in its own and it's in its own right. It's a rather popular rigorous method and They've they've used it here that the gist of it is is that as opposed to having kind of a flat level of Criteria which then is aggregated to derive a alternative value The criteria is grouped into subsets Into a sort of a logical tree structure hierarchy to develop intermediate levels of criteria Valuation you can see this in in the it's a rather busy slide But in the center there I have a large blue circle around soil So you see there's a number of discrete criteria for soil and then those are Rolled up together into a soil suitability Criteria valuation by doing this first of all you simplify the model, but also avoid some Overweighting of criteria that is related to each other in the case of their work here after arriving at a theoretical suitability then they really used a hybrid approach where although they didn't call it out as such It's a reference point method where they've then looked at their theoretical suitability and eliminated Alternatives that don't meet certain constraints that they've applied In the next example, this is the sighting of housing development in Switzerland Using an outranking method and what I found interesting about this work was that well without ranking you have to Identify specific alternatives, but in many sighting decisions you theoretically have an infinite number of alternatives So to work around that issue They used a closeness relationship methodology to identify homogenous zones which would represent Discrete alternatives that they could then take through the outranking methodology to derive a Sighting map that indicates suitability as favorable unfavorable or in the case That is a possible outcome from outranking Uncertainty about the suitability So I've I've gone through about 30 different GIS based MCDA method descriptions and case studies and I've summarized them here I found predominantly in the work that I've pulled up that value measurement and specifically analytical hierarchy process methods were predominant I've also indicated the types of decision problems and Most of these were associated with land suitability and site selection There was a rather comprehensive Literature survey done by Yossiq Malzewski in 2006 in a highly cited review article where he did a lot of categorization. These are the results From his work. You see this matrix of decision problem and application domain so he identified that that the most predominant the decision problems were land suitability scenario evaluation site selection and resource allocation and application domains included environmental applications urban planning forestry transportation hydrology and Waste management, but there were was quite a distribution and so it gives a flavor of the breadth of the application of GIS based MCDA So now let's talk about Retrospective GIS based MCDA This is something that I have not come across in the literature, but intrigue me I put forward the hypothesis that given a large enough population set of similar historical spatial decisions That we can take an inverse problem approach to determine the subjective valuation of criteria from the stakeholders without a priori knowledge So in a traditional forward theory we take Parameters we plug them into our model and we derive a predicted result an inverse theory You take observe results assume a model and Work backwards to try to come up with the model parameters and I intend to apply that to a prior GIS based MCDA decision set So this is the summary of my capstone. I will integrate GIS and MCDA to retrospectively examine a prior site decision case study data set and I will look at a case study which fits the bill for the type of problems that come up with MCDA that have Multiple stakeholders conflicting motivations and uncertainty in the data My approach will be to actually look at the decision results for a domain case and contrast Predictive results using regression and stochastic analysis to to look at criteria waiting and and its uncertainty Without any explicit information about what the stakeholders actually did to come up with that decision I hope my objectives I hope to be able to create a probabilistic model for prediction of future related decision outcomes and Provide insights really trying to get inside the head of the stakeholders. So get insights in their strategies and This is a novel methodology. So I hope an outcome of this is that I can Demonstrate this is a new methodology applicable to other GIS decision domains So I had to choose a case study topic This is Really of course I an MCDA problem in its own right, of course I start off with areas though that You know that were familiar to me from my work experience particularly in the energy and environmental field So I I really consider a number of different potential topics But made my selection really based on the availability of the data because it's important because I expect them going into new territory The statistical analysis I want to ensure that I have a large population set that there is some consistency in that set then I can get at the source information readily and result of this selection process I Arrived at selecting waste transfer stations as my case study domain The sighting of waste transfer stations waste transfer stations are Intermediate points for solid waste it comes from its source and residences or businesses Ultimately it ends up in a landfill or an incinerator But it's not always cost-effective to deliver it from the source point to the final Destination because of the cost of the small vehicles that drive around and pick up people's municipal waste So often waste transfer sites are located at an intermediate point representing the optimal location So that the solid waste can be in some cases sorted compacted But then transferred into a larger haul vehicle now, of course waste transfer stations are on the one hand have to be near the source where people live But no one wants them near where they live So it's a good NIMBY problem and has a lot of conflicting Issues associated with sighting. There's a lot of information available on best practices There's some EPA manuals and documents that I've shown there. There's also Organizations that deal with this so it's it's a I find it's interesting. It's a field in its own right But having best practices doesn't necessarily ensure that's how decisions were actually made And that's what I'm going to dig into Also with these decisions and one of the complexities that there's many stakeholders and because of the NIMBY issues, of course It's a community and neighborhood issue Sometimes the waste transfer stations are commercially owned and operated. There may be industry involved Environmental organizations of course local and state officials and public work officials and sometimes academic institutions So it's a ripe MCDA GIS problem Waste transfer station permitting varies from state to state. So I wanted to stay within state boundaries and Looking over the large state Arrived at California because of its large population large number of waste transfer stations And also a good database that has a lot of transparent information about their waste transfer stations There's over 3,000 solid waste facilities identified in California's solid waste information system database 700 of those are active waste transfer stations And then I pared that down to about half of them deal with mixed municipal waste, which is what I'm going to focus on In addition to having a lot of information about the waste transfer stations They also have linked access to the permits, which I'm going to have to dig into to understand when these were built In other details that are going to lend a deal that I'm going to have to deal with in looking at criteria Uh, so as an example here, I'll zoom in on orange county. There's 41 waste transfer stations in orange county They serve three county landfills In the middle of this slide, you'll see the area. I've shown one waste transfer station the CVT regional Transfer station. It's privately owned and operated for the city of Anaheim There are a number of other waste transfer stations within a three mile radius fit So they're effectively competing for the same solid waste and that waste then is delivered to the linda alpha sanitary landfill About 10 miles over the road Over on the right hand aerial is actual You can see a bird's eye view of what that waste transfer station look like. It's a covered facility Ready access to the highway in an industrial zoned area But it is fairly proximate to a residential area with townhouses and single family dwellings just across the highway Now one of the challenges I'll have just I'll contrast that with another Waste transfer station in western california. So I don't have to describe too much about the details of these waste transfer stations for you to understand There's substantial differences that went into the siting of these two locations The the one on the on the right is in the San Bernardino county It's actually operated by the california transportation department co-located with a maintenance facility And doesn't quite have the same kind of NIMBY issues. So I'm going to certainly have to go through this this database and identify subsets of similar types of facilities so that I can have a reasonable opportunity to be able to do good statistical analysis Here's a an overview of my methodology. There is no pure inverse problem because you can't just take Results and work backwards without any knowledge of what happens. You really have to come at it from both sides So I have to really go through the problem structuring phase and then take the results and converge on the model building trying to identify a useful model and develop Through statistical analysis the parameters that went into the criteria and its aggregation I've completed our preliminary data preparation and in the process of reviewing and deep drilling several site locations Where I'll actually go into Beyond what I will be able to do with the broad Set of data that is get into the public record and understand exactly what transpired when these were cited Then with that information, I'll prepare the the balance of the data It's going to require some automation techniques to pull up information from permits And then break out some homogeneous subsets that I'll work with I'll develop model structures I'm going to focus on value measurement reference point decision rule models because they're It's more logical to reverse engineer them and then applying both Deterministic regression methods and stochastic analysis. I will try to derive Criteria parameters partial values and the reference points for these data I'll evaluate the model effectiveness goodness of fit To try to understand what the uncertainty is in in the model and then as outcomes I hope that I'm able to characterize these site decisions In the form of a probabilistic model that actually would serve to predict future decisions And with this also with this Criteria valuation gain some useful insights into the stakeholders own subjective valuation process With this I'll have performed an assessment of this new retrospective GIS MCDA methodology I'll learn what other kinds of insights can be gained in terms of stakeholder strategy And the predictive effectiveness of the method as well as its deficiencies and development needs I will assess the methods of men ability to other application domains and other GIS decision problems and recommend future work requirements and practical applications It's my intention to publish the results of this study in a referee journal in addition to Presenting at one or more conferences So this is my timeline I plan to Complete this work in the spring two semester So I'm on a rather short fuse to submit an abstract with the intention of being able to Make a presentation sometime during that semester ideally in april But I also will be preparing journal paper and submitting it Towards the end of the semester The the body of work I described in my methodology then it's underway now And it's going to be rather intensive for the next couple of months I will largely have to have completed that work by april I've poured over Dozens of articles about GIS MCDA and I kind of feel like I've just scratched the surface There is a lot of rich information out there. I learn a little bit every time I look at something It's going to be an ongoing process I'd like to acknowledge a few people Two individuals from Arizona State University professor Tom seager and Valentina Prada Who I've worked with on MCDA as it relates to sustainability analysis And and that was the inspiration for looking at it and from the perspective of a GIS I'd also like to thank Doug Miller who I started conversations with this going back to last summer And got me started for the semester and justine who took over afterwards and has been Very very helpful and who I will rely on very heavily as I get into the rigorous statistical work in the next couple of months So with that, I want to thank you and I'll be happy to take questions Okay, I see some folks typing Dianne asks will this model be developed as an application down the road? That's a little bit hard to imagine right now. I think it It's a possibility That seems to be the route for a lot of these MCDA Methodologies they start off as concepts and methods and and many of them have Developed into applications. There are also some Some application toolboxes that have a lot of different pieces to them one of them Discussives to both Doug Miller and justine blanford is adresi Which is a toolbox that does a lot of things with GIS including MCDA so it's possible, but I'm not thinking that far ahead yet. It won't be an outcome of my work this next semester Thanks regarding the peer reviewed articles There's a an individual I mentioned his name yasik melzewski. He's at western ontario university And he's kind of been the guy who's done Much of the survey work through a number of papers. He actually wrote the book. It's sitting in front of me It's called gis and multi-criteria decision analysis Uh, it's the only book on the subject and he's about ready to put another one out So he did the the footwork on that 2006 survey and I expect he'll be doing update work on that as well With his new book that'll be coming out probably just about the time I finished my work Thanks to him. Okay, perhaps that's that's it Oh, we'll give it a second here to see if there are questions there No, thanks Thank you. John great talk john and and again MCDA is something that is has been overlooked in a lot of quarters and it is a It's a really interesting way to be able to to think about how do we quantify decision-making and you're looking forward to To your work. It is an aggressive schedule, but I think you're set for it and ready to jump into it and We'll check in with you at the end of the end of spring two Um, josh, okay. Thanks, uh, duiz is our next speaker and josh if you want to go ahead and load your You can get started whenever you'd like josh Other likes starting at the end and going back to the beginning um Good evening. I'm josh duiz and my project is Um regarding utility vegetation management application I'm going to be designing Uh, mobile geospatial application to streamline some field collection processes In my current job and and maybe improve the record keeping that we currently do I'd like to introduce this presentation with just a quick Quick interview or overview of the topics that I will cover I want to start by laying out a few goals for the design. I'll talk about what makes a good utility vegetation management program I'm going to cover a few basic operations of the distribution utility And present a case study of the vegetation management program here at spoon river electric cooperative And then from there I'll move on to talk about my prototype design strategy and the basic information products that are required And a proposed timeline for this project After that I'll open it up for any questions The goal of this project is to design a prototype for a mobile application That can be used to streamline the vegetation management field survey work here at spoon river electric cooperative This application will improve the consistency in quantifying vegetation management work And improve the record keeping and documentation of the work performed by the crews I put a few photos on this slide to use as examples of vegetation management work You'll see uh some manual work here and uh a right-of-way mower that we use And some bucket trucks here doing some side work In the right-of-way areas Vegetation management is one of the most expensive maintenance activities in the electric utility business It's necessary for safe and reliable electric grid Safety and reliability are two things that will cause public action and are mandated and regulated by government agencies Maintaining vegetation requires a dedicated program that uses a solid planning strategy This can be based on certain time schedules or a rotation It can involve contracted workers or dedicated staff But it requires a consistent policy and budget to be effective It requires detailed records to track the spatial characteristics over time The important factors are tree species Volume and location Tracking this helps to identify the best integrated vegetation management strategies over time Here are three main ways that vegetation impacts primary distribution lines direct contact Overhanging branches and underbrush that prevents access to the line for inspection and maintenance activities This is just a photo that shows an example of what happens when trees make direct contact with high voltage bare conductors The area inside that yellow diamond shows where vegetation is self pruning by burning off at the wire This causes a broom effect that you can see clearly in this photo It also causes flickering lights and outages Yellow diamond here just indicates where the wire is this photo. It was a little bit hard to see Where the wires were going through there Um The branches over hanging the wire can be a major cause for outage during storm events that include high wind or freezing rain And this time of year in illinois We often have periods of freezing rain that can cause branches to To become heavy and fail during wind wind events and when they break off they interrupt power either By physically breaking the line or activating the protective devices This picture includes some trees growing directly Under and around the lines that can definitely impact not only power quality But also the access to the line if it happens to break or come down If maintenance activities are required Even though this is right along a county road still would pose a very difficult challenge to get back in there and repair that There are a few basic things to understand about the electric grid and how vegetation impacts it Faults occur when vegetation provides a path to ground for electricity flowing through the power line These can be high current or low current events High current events can cause outages by activating overcurrent protective devices The picture on this slide Are examples of two basic system protective devices The picture on the left is a fuse cut out And it works very much like a a fuse in a fuse box at home When the current becomes too high It contains a sacrificial sacrificial braided fuse link that burns through And interrupts the flow of power The picture on the right is a simple oil circuit recloser This acts like a breaker that can reset itself automatically. This allows transient types faults to clear Before re-energizing the line. So if you have a branch that falls into the wire Sometimes it'll go ahead and fall down through and then this will open long enough to let that Fault pass Can everybody hear me now? Okay, I'll just start over here with this slide There are a few basic things about the electric grid And how vegetation impacts it Faults here Can occur when vegetation provides a path to ground for the electricity flowing through the power line They can be high current events or low current events High current events can cause outages by activating the overcurrent protective devices These are fuses and oil circuit reclosers A fuse basically acts like a fuse in your fuse box at home it Over when it when the lines overloaded it'll activate blowing the fuse it'll open the the trap door and disconnect power A recloser acts like a breaker that can reset itself. So It will open temporarily allowing a transient fault to pass pass through the line such as When a branch falls into the wire It might cause a high current event that will cause the power to disconnect allowing the branch to fall through and then It'll close back in and power will be restored with just a momentary interruption These issues cause customer dissatisfaction economic disturbance And ultimately regulatory issues So proper vegetation management minimizes the number of faults caused by trees and animals Spoon River Electric Cooperative operates a vegetation management division They perform vegetation management work for three rural electric cooperatives here in central Illinois The graphic shows the general location for these cooperatives Spoon River has about 15 employees whose primary function is to perform vegetation management activities This program currently uses entry software and handheld GPS equipment to plan and document the work activities There are several manual steps to transcribe data from the field into the GIS database The current GIS model Utilizes linear referencing and route event tables to create paper work maps for the field crews The field work is then documented in a similar way where GPS points and paper work logs are manually recorded in the database These can be Sources of error or places where transcription errors can occur This is the current workflow for the vegetation management program The planning process starts with the job foreman and myself Performing a pre-work inspection of a substation This involves traveling the entire route and documenting work areas with the Garmin 72h GPS Handwritten notes are taken regarding the type of equipment required the access constraints steepness of the terrain percent canopy coverage and special notes such as notes about member concerns or access issues The GPS is downloaded in the office and paper notes are transcribed into the GIS database Color coded maps are then printed from that GIS for the field crews to use When they go to do the work After the work is performed the field crew collects GPS points and documents what was done and when it was done Also any customer issues that come up can be recorded during that The worksheets are collected weekly and this data is also transcribed into the database for long-term record keeping This can be used to calculate work footage and operation reports This is an example of the field sheet that we use for data collection of the pre-work inspection This sheet was developed over time working with the the job foreman To make data collection simpler and more consistent One of the places for error were in handwritten notes the more variation there was between different job foreman the more issues I had in transcribing that into the database to get an accurate picture of what was required in the field So that was a place to introduce transcription errors and misinterpretation of those field notes This is the current database table that we use and the database reflects many of the same fields as the data worksheet From the previous slide It includes additional fields used to calculate the footage and work units needed for estimating and budgeting Calculation used for this is shown here at the top And it's very simple, but it's really improved the our ability to estimate accurately how long it's going to take And what's going to be required at the job site So just having the consistency of the work unit Didn't didn't give you a hard-coded value For what that working it was going to cost or how long it was going to take But it allows you to kind of make an interpretation based on the crew whether they're capable of accomplishing 10 work units in a day or 20 You can at least evaluate how long it will take to do that work based on any given crew So this is a sample of the field map That we would typically print out for the work crews This along with the written notes that are taken are used by the foreman to identify the work in the field Yellow indicates a bucket tram green might be Some sort of manual work and the orange might indicate the need for the right-of-way mower In this case there might be There might be multiple Lines and this indicates that you're going to need multiple crews to perform that work So the goal of this project is to create a prototype for a mobile application that streamlines this manual process This will improve the efficiency in both the planning and the record keeping steps of the workflow It would also reduce the need for print maps For the field crews and these could be replaced by the mobile version so that updates appear as the changes were made To start the process. I have a workflow diagram showing my anticipated workflows I'm using the case study here of spooner over electric cooperative as a starting point And from that I want to design a survey that will go to all the distribution cooperatives in Illinois The purpose of the survey will be to identify vegetation management strategies at these other cooperatives And this will be used to Identify any additional design criteria The initial prototype will be tested here at spooner over electric cooperative In our vegetation management program due to time constraints for this project I realize it may not be possible To do all of this within the the scope for this particular educational project But I'd like to continue to develop this after that point The survey will be sent out to the cooperative managers in Illinois and this serves a couple of purposes First it gets the upper management involved and lets them know why i'm asking questions About their private business Um and the manager is also the best person to talk to to identify Who who's actually the boots on the ground? That I need to talk to regarding the more detailed questions about the work methods I also want to take a look at what is available From commercial off the shelf solutions And so discussing that with all the other co-ops. I feel like if There's anything out there that they're currently using that I can evaluate that might be a good idea to to get that feedback from them As well as any research that I do on my own So another big consideration for me is the server implementation Options include the cloud services such as our gis online and and amazon aws And using Even the traditional rgis server that I have here behind my firewall to publish to one of those services Uh will be considered Just deciding whether to use Cloud technology or whether it's better to just use our existing technology will basically be a financial decision um I think the cloud makes a good case for the web server for sure And using a tool like our gis online While the main focus for this design is on the features Of the application another important aspect Will be the interface that will be used for data entry Certain constraints will be included to maintain the consistency in the database And these include features like coded domains and some some boolean fields Um, and they can be accessed through through through different types of interface like drop downs checkboxes radio buttons And this paper prototype Will help me identify a user focused preference For those things, uh, and I would like to work that out in case there's custom programming that I need to do To create this this type of interface through the web application My initial design requirements include a mobile design that works with tablets or smartphones And this application needs to be capable of disconnected editing and and capable of synchronizing back with the distributed database When a network connection is available This application also has to be very easy for my employees to use in the field Uh, as this will be a change in the workflow for the current field personnel If it's not if it's not easy, there'll be resistance from them probably to adopt it and Uh, that'll be a problem Moving forward with the application if if they're not using it It'll be hard for me to make a case that it's something that we should move to Uh, I think the most important aspects of the design process Are identifying the primary information products I identified these Products that are listed here from the current workflow number one Create maps for communicating the work type and the location to field crews. I think that is the the primary goal Uh, number two create records showing the daily work process progress and the long-term maintenance records And number three creating the estimates for the work units and describing each substation or circuit This list really Boils down the primary functions of the current system So product one is maps for communicating the work type and location to the field crews This product should answer a job performance question. Where are the trees and what equipment will I need to clear them from the right away? product two is about records Showing daily progress and long-term maintenance history and this is really meant to To answer somebody in my position the line clearance manager's question of is the work complete When was it complete? And were there any problems or concerns from the members that need to be documented or dealt with and product number three Estimates for work units describing each substation or circuit. This really focuses on what does the ceo need to know? What does the board of directors want to know about this? And it's how long should this work take? How much will it cost? And how are we performing against our budget and our work estimates? Without that answering those that last question, uh, it's uncertain whether funding will be available to to actually produce this this product So my capstone timeline I propose this timeline of approximately six months Kind of the seventh month will be used to review the results and develop my conclusions from the design process I anticipate presenting this at the edge or user conference in july And I hope to have a working prototype by then to show at the conference I recognize that this may May be a little out of the scope for the design proposal Uh, but I hope to have that ready By july to present this work And this is just a a gantt chart showing my timeline It kind of breaks the work into manageable pieces and helps me stay on track for my presentation So I have a baseline at least To gauge against as I go through this I'd like to thank dr. Anthony Robinson for Uh serving as my advisor and first patients and assistants in this endeavor And I also have some references if if anybody's interested the last couple slides are Are my references, but they'll be posted later If you want to see them if there's any questions, uh now would be the time for questions It says sounds like a very useful app. What device is it aimed at? It probably be Aimed at the iPads at this point just because we have those available to us at the co-op There's also a lot of uh, a lot of the guys now have their own android phones and so Probably be a balance, but I suspect the ipad will be a little more user friendly just because of the size Asking about including pictures and yes, uh, there's definitely A gps units that are capable of Pictures there's also pictures from the ipad and using the gps in those units And and those are definitely things that I would look at probably out of the scope for this This part of the project, but uh Definitely in the future. That's something that we would want to include Uh No, probably not native apps at least not right at the beginning web based applications Um the stuff that's available through arc gis online It's probably uh more what I had in mind to begin with at least to get through the prototyping You consider letting users design their own interface from the paper parts and yes Actually, that's that's part of the Idea behind the the paper prototype testing for the interface It's to let them make some suggestions leave it a little more open-ended so they can Describe what they what they like Does this need to work in all weather or using gloves? Ideally it would be better if you could Use it in any and all weather conditions, especially in our business And be able to use it using gloves, but I think A lot of that record keeping can be done from inside the truck As far as Filling in the blanks So that's that probably won't be part of this first prototype anyway Uh, yes, that's what uh, I've been looking into you the the especially as every Um generation of arc gis comes out. It seems like they have more to offer In uh technology and one of the things is the offline ability to have disconnected editing even on their web applications And so that's definitely a concern for me because there's still a lot of places where You can't get uh cell service in the areas where we're working If if the majority of other co-ops are using off-the-shelf solutions I guess that means that I should look at those same solutions I'm somewhat confident that at least the co-ops here in illinois aren't doing that Mainly due to expense I'll be interested to see what I learned from those other co-ops because That's definitely an area where I don't have a good answer So if there are other off-the-shelf solutions, then that's what I want to look at Are there any other questions? Okay, josh, uh, thanks a lot the great Great project in the in the sense that uh, moving the moving getting your workflow and And essentially streamlining your processes for something as as This is this is every day you do this stuff and those those kinds of applications Are are worth their weight in gold We'll be really looking forward to see how this comes out when we're out at the user conference Our our next speaker is chris dunne and chris You can load up your slides whenever you get a chance I'll remind everyone again that we All of our speakers value your feedback and if you weren't able to ask a question Or if you think of something later, please use the url in the upper right corner there to To to provide feedback that goes directly and immediately to the student and they can they can see those those comments Also remind everyone that we'll do this again tomorrow evening and again on thursday Excuse me and again on thursday evening And uh, we'll have uh, I believe we'll have a total of 11 talks Uh, uh this uh This fall two Um Semester and chris i'll let you get started whenever you're ready to go and we'll learn about your work in portland All right. Thank you. Good evening everyone. I'm chris for dunne My project is a quantitative analysis of the bicycle infrastructure of portland main, which is where I live My advisor is dr. Alexander couple Um, so here in portland It really feels like bicycling is becoming more and more common. Um, as a lot more people move towards more sustainable lifestyles And bicycling is frequently used by many people as a way to commute to get exercise Or to simply just get outside and explore Um, there are multiple benefits to bicycling including like personal health benefits And benefits as far as environmental sustainability And uh, but despite those benefits, uh, many people are concerned about safety when riding a bicycle And a lot of people can be deterred by the risk of getting into an accident, especially one with a car. Nobody wants that Um, not only has cycling become a more mainstream activity for the general public But government guidance documents are including cycling guidelines at both the federal level and at least here in main We have state specific recommendations as well So at the federal level The united states department of transportation has a specific policy statement regarding cycling and pedestrians Basically, they acknowledge that infrastructure specific to bicycles or pedestrians Is equally as important as infrastructure for vehicles Then they have minimum recommendations for design features to maximize bicycle and pedestrian safety Not only do they have those Minimum recommendations, but they recommend that municipalities and transportation planners and other agencies Go beyond the minimum requirements as a way to ensure that long-term transportation demands can be met And at the state level Maine has the sensible transportation policy act, which I think is just a great name It requires that all options including things like improving bicycle infrastructure Be considered before a municipality Simply improves their vehicle capacity with more lanes or wider lanes or higher speed limits or something like that This policy also provides incentives for non automobile based transportation planning And at the local level around portland There is the portland area comprehensive transportation system Which covers portland and several surrounding towns and they have a regional transportation plan Which has some even more specific details about cycling and it calls for improved and expanded bicycle infrastructure Both to help with safety concerns and for sustainability Really in both the federal and the state cases The policies and recommendations are not just designed for safety But it feels like they're actually there to help planners Find ways to encourage people to use alternative means of transportation As a way to curb vehicle traffic growth and to help deal with population growth in a more sustainable manner So when I talk about bicycle infrastructure What I mean or what bicycle infrastructure is it's Physical characteristics of a roadway that are intended for use by cyclists as a way to better integrate them with automobile traffic There's many different types They can be defined in a few different ways, but some of the common ones you'll see Include bike lanes, which probably everybody has seen it's probably what most people will think of when I start talking about bicycle infrastructure This is a delineated lane intended to be used by cyclists That's part of the regular roadway Another one you might see is a paved shoulder Which is just a small area to the right of a travel lane that you know is marked off But it's not really bicycle specific and there may or may not be parking in it or bus stops as you can you know see in this picture There are also shared lanes which are frequently denoted by shared use arrows Which are the ones you can see Right here Those are also called sharrows They tend to be used in situations where the road is too narrow for a paved shoulder or a bike lane And they are frequently accompanied by signs that say share the road Or something indicating that cyclists have the right to use the lane as well as cars You'll also see mixed use paths which are physically separated from the road They provide alternate routes for pedestrians and cyclists This picture here is a very popular path in portland called the bat cove trail Which connects many dense residential neighborhoods that are off of the portland downtown peninsula And it connects them to routes which lead downtown While also avoiding major arterials. It's a very popular path You know, I've walked around it many times and there's always lots of cyclists and it's popular for Recreation and running and it's just really pretty in the summer as well You'll also see bike boulevards Which are slower smaller local streets where cycling is encouraged as a way to avoid more dangerous roads So sometimes they'll plan To indicate that specific side streets or bike boulevards to help cyclists avoid specific areas or specific roads So roads that have low amounts of traffic and low speed limits Can be given the additional signage to help cyclists, you know, find their way along the boulevards And another major type is the cycle track Which is a bicycle only lane that is physically separated from the road They tend to be highly favored by cyclists. They're fairly safe And we don't actually have any here in portland I think you probably see them more in major cities because they require a bit more infrastructure And you know a lot more work to put in and you need a wider street or a specific situation where you can actually You know have one installed This is the current bicycle infrastructure here in portland Um, I will show a few specific spots. There's This road here is bright and av and that is one of the major routes into the city And as you can see that is a shared lane. So there are shadows marking that you know along along the road And then this route here is forest avenue, which is the other major route that a lot of cyclists and a lot of cars use It gets very heavy use and that has some bike lanes But then there are spots where the bike lane stops and it drops down to a shared lane Because of just whatever the situation is or how the road was designed and because the roads were there For a long time before they added the bike lanes Oh, and there's also the backcove that I mentioned. That's this right here You can see the multi-use path goes around the cove and the road on the western side of the cove Actually has a bike lane as well So earlier I mentioned that cycling has quite a few benefits There are a lot of documented benefits as people have done research into this The personal health and fitness benefits are kind of the most obvious that a lot of people might think of Cycling can burn a lot of calories and regular exercise helps with cardiovascular health and can greatly reduce rates of obesity and related ailments like diabetes Uh, financially if you cycle as an alternative to using your car such as for commuting or to run errands You can save money on gas and car maintenance and depending on where you live You might be able to give up owning a car altogether And as far as sustainability goes, um, choosing a bicycle over a car Will reduce air pollution And noise pollution and lessen the consumption of fossil fuels One study analyzed the various benefits and risks of cycling and found that the benefits are substantially larger than the associated risks So what are some of those risks? The first thing most people think of that I mentioned earlier is personal injury due to a crash That could be a crash with a vehicle or also a pedestrians other cyclists or A stationary object like a fire hydrant or a signpost While there are the health benefits, there are also the health risks other than crashes or personal injury Increased exposure to air pollution is a major one, especially fine particulates emitted by automobiles Particulates are linked to respiratory problems like asthma and other conditions of the heart and lungs One study actually found that car car drivers are exposed to a higher concentration of particulates But because of increased respiration during cycling Cyclists will inhale more air. So their lungs absorb more particulates than if you're driving a car So presumably the problem of air pollution exposure would be lessened as more and more people choose walking or biking over driving a car But despite these risks the health benefits of cycling alone have been found to have a net positive benefit to personal health And then the additional societal and sustainability benefits on top of that from bicycling are just kind of icing so As far as bicycle use rates go the usa is fairly low um, you can see in this graph about one percent of Trips are estimated to be taken via bicycle Some other countries are much higher than netherlands is a pretty common one that most people reference there at around 26 percent There also is a study that has looked at rates over the past 40 years or so which interestingly a lot of the country these countries have stayed relatively Stagnant or have even dropped off a little bit. So at least the usa has not dropped We've actually raised it a teeny tiny bit And other countries there's a lot of different factors that go into bicycle usage the netherlands in particular Public opinion is very different. The infrastructure is very different. Their health system is very different So there are a lot of things that go into into these numbers And even with the higher per capita ridership in the netherlands Fatal accident rates are very similar between the us and the netherlands, which I thought was interesting And there were both roughly at around two per a hundred thousand A hundred thousand people with riders So yeah, but like I said a lot of different things go into that And there are two Specific studies that I wanted to highlight. So while researching this project. I was steered towards these They're frequently cited in the world of bicycle research and advocacy One is safety in numbers more walkers and bicyclists Safer walking and bicycling is the title which you can probably guess what that's going to be about Um, and the second one is route infrastructure and the risk of injuries to bicyclists a case crossover study by k tesky So the first study by p. L. Jacobson Basically they looked at the relationship between the number of pedestrians and cyclists and the rate of accidents involving cars And those pedestrians and bicyclists They had several data sets for a variety of countries And they wanted to find out if increased bicycle and pedestrian traffic actually resulted in more accidents between cars and non-cars Because a lot of people seem to think that that would be the case Um, what they actually found was a strong inverse relationship Meaning that as the amount of pedestrians and bicyclists relative to automobile automobile traffic increased The number of accidents actually went down. So in other words If more people are walking or cycling in an area Then it is less likely that they will be injured by a motorist and the safer things become for pedestrians and cyclists Which I think is a pretty important conclusion as far as the world of you know cycling goes And the tesky study Compared the relative risk of injury associated with various types of bicycle infrastructure They surveyed several hundred people who were injured in cycling accidents to find out the details of their route And then they compared their their routes The routes were injury occurred To randomly selected routes with the same start and end point as a control And then they calculated the relative risk of each type of cycling infrastructure using regression analysis And this is a figure from the tesky study They have compared the risk associated with each type of bicycle infrastructure To route preference which came from surveys of cyclists done in another study And they've highlighted the infrastructure types that are both viewed favorably by cyclists And are considered more safe based on the results of their study So here you can see the cycle track Which as I said we don't have in portland is both highly favored and considered very safe based on their data There's also the bike only path here local street with a bike route A local street without a bike route or a major street It's major street with no parked cars and a bike lane. Those are the Types of infrastructure that are the most highly favored Highly favored and considered the safest by the study What I think is important about both of these studies is how They show the importance of increasing ridership As far as safety goes and making planning decisions based around both perceptions of safety And empirical safety data Because even if a type of bicycle infrastructure is considered safe by the people who have studied these It may not actually be utilized if the public doesn't see it that way And if we want to increase ridership to increase safety, we need these things to kind of happen in tandem So I'm going to go over the data that I've gathered in order to do this project This map shows the locations of accidents in portland involving bicycles and cars This just covers 2011 to 2013. I have been on the hunt for more data And actually about four hours ago. I finally got an email with some crash data covering 2004 to 2013 So hopefully that gives me a much more robust data set to work with So in order to start developing an understanding of what may be the driving factors in portland's crashes I started looking at different attributes that came with the crash data So I have time of day Which is broken down. We've got daylight dawn or dusk and then at night either in the dark with a light or without a light Road conditions most of the accidents appear to be on dry roads, which I actually was kind of surprised about But there's also some on sand slush or wet roads Then I also have information about the type of locations such as a curved road a straight Straight road of various types of intersections or driveways Whether when the accident took place whether it was clear cloudy raining, etc I also have more quantitative data associated with each roads with each road segment that I could join with the crash data To perform an analysis. I've got the speed limits, which are shown here And the average annual daily traffic as well I am looking for additional data The accident data like I said, I just got some so hopefully that helps out a lot The other key piece Is bicycle ridership data. We want to know how many people are actually riding their bikes on these roads Having those counts will let me compare the number of cyclists relative to the number of cars Which I think could add a lot of depth to the analysis There are two automated stations that count bicycles along that back cove trail But I haven't actually been able to get that or those data yet But I would really like to see bicycle counts on the actual roads with the cars The greater Portland council of governments, which is one of the organizations I've talked to a few times during this project They have some count data, but it's very sporadic. I'm still kind of looking at that to see if it'll be useful They're also planning on purchasing ridership data data from Strava, which is an app similar to map my ride It's an iphone app and You people can track where they you know where they start and end and they have a lot of data available This map shows the data that they have on their website that they they only show it for portland You can't download anything, but they have a lot of coverage. So I think that would be really Interesting and useful to look at I could also calculate a few things to add into the analysis such as the distance to intersections at each Accident location or accident density along road segments Or population density. So there's a few different things that could be involved So for the analysis We're looking at a few different ways and sort of assessing still assessing the data to see what sort of analysis will be most appropriate We can start with simple visualizations looking for spatial patterns like a hot hot spot analysis or heat map We can use statistical method methods such as chi squared test to look for dependence using the qualitative variables Like the time of day and the road conditions and the bicycle infrastructure, which is of course one of the big ones I could also use the quantitative variables like the speed limit the traffic Or anything that I might calculate on top of that for a regression model To start really looking at what Might be most influencing, you know or causing accidents to happen here in portlands So my anticipated results, I really just need i'm looking to find or get a better understanding of factors that are contributing to the safety You know either increasing or decreasing safety for bicyclists here in portland And beyond actually competing the project, I would like to make recommendations to the city or to the the gp cog for Steps that could increase the safety of the current bicycle infrastructure I don't or to assess the safety of any proposed bicycle infrastructure I've also thought about maybe making a web map like a safety web map showing Routes that people might want to avoid or really safe areas or high crash rates or things like that Here's a map showing the proposed infrastructure for portland and you can see there's Quite a bit of shared lanes that they're hoping to put in in the downtown area Along with some bike lanes leading downtown off of that backcove path So my timeline the next two months or so I want to continue to gather and assess all the the data And then I'll you know conduct the analysis write my paper and I've submitted an abstract To the american association of geographers conference in chicago at the end of april so I will be presenting something there And I just have my references page which people can review later if they like And then just a few thank yous to my advisor alexander clipple Rick harbison the transportation coordinator with the greater portland council of governments Jim tacit the bicycle coalition of main who pointed me to some studies and to rick and Dwayne brunel who just sent me that crash data From the main department of transportation So I think I still have a few more minutes if anyone has any questions So dn says have I considered making a generalized model that can be used by other cities? I think that would be great. Um, I definitely will sort of depend on On my results and I think it will be interesting to look at because portland is a fairly small city And we have limited, you know bicycle infrastructure. So I don't know If the model would be that scalable, but it's definitely something I would want to look into And I think other people have done similar things in some bigger cities Most of the studies I looked at were actually from canadian cities So i'm not sure how much has been done here in the us Matthew says where you're getting the detailed information on road conditions light conditions that Um, you're right. I think that is unusual for bicycle accident reports This all came from the main department of transportation and I believe they are getting it from the police database So that is actually I guess that would be considered Uh, some bias in my data is all these crashes are only ones that actually ended up getting police reports So if it was a minor altercation or that nobody called the police then Um, it wouldn't end up in this database. There is a voluntarily reported Bike accident database that the bicycle coalition of main has that hopefully they'll be willing to share with me They just have to scrub Uh, some of the personal information out of it So hopefully that can supplement what I have from the police department and from the main dot But it might not have some of the same Attributes so I don't know how well I'll be able to meld the two Dian says is Portland perceived by cyclists cyclists generally safe or is their community-wide unease? I think it is considered fairly safe Um, one of the biggest issues are those two main roads that pointed out for stav and brighton av They really are the best way to get to certain parts of the city and they're not really great for cyclists They're just, you know, they're very old roads. They're not always that well maintained the infrastructure isn't great I'm a member of a couple bike commuting groups on facebook and on meetup and you know, the people definitely have Some issues but for the most part I think portland is considered a very bikeable city even though it's really hilly so it gets difficult sometimes Nicholas says Toledo, Ohio has a well-supported public park network, but I don't know how many bike riders have ever seen There might be a good possibility for future comparison. Yeah, that's that's true The back cove trail is actually maintained by an organization called portland trails And they have quite a few trails around the city some of which They encourage people to bicycle on And those some of those might not actually be on the bicycle infrastructure map. I have so I should look into combining those Or seeing if if they have any idea of how many people use their trails for cycling like to avoid specific roads other than using back cove Yes, though, uh, john says plenty to do some field work on the subject. Unfortunately between now and april It's not really prime cycling weather here in maine But I might have to you know ground truth some results next summer All right. Well, thank you everyone. Um, I don't it doesn't look like anybody else is typing Thanks chris I know that this subject is near and dear to your advisor's heart as a as a european who um enjoys his bicycle and Finds it strange the way we Failed to have connected with that device in america. So looking forward to your results out of the age He has mentioned that once or twice Yes, just a couple times alex wood um Okay, our next speaker, uh scheduled to be matthew white He's actually in uganda So it's about uh, if I if I'm correct, uh I think it's probably three or four in the morning there um and He doesn't seem to be with us. Um First one of the first times this has happened, but not surprising that uh in a developing country. He may be experiencing Um the inability to to get on the internet We're going to work at tracking that down for either one of the next two evenings to see if if we can Get matthew connected and have him give his talk um We'll start to work on that as soon as we can and um We do have a uh Um, we do have a time slot on the last evening that is unfilled. There are only three Uh speakers that evening so um, I'll Let everyone go at this point and uh, please tune back in again tomorrow evening We will um begin again at 6 30. We'll have uh four speakers tomorrow evening um And we'd also ask you remind you again to consider providing feedback to the speakers through the Evaluation form at the url at the top of the notes pod I want to take this time to thank all of our speakers tonight. Um great projects Uh Really just shows the breadth and depth of our program And the quality of our students and uh the impact that they make In their everyday jobs and uh in their communities So I want to thank you all for that and i'm very proud of the work that I saw tonight And I know that tomorrow evening we'll have a We'll have another set of equally good Equally good talks so Um good evening and uh, see you again tomorrow evening at 6 30 In the chat room here Good night