My name is Andrea NORRIS And so they're presenting with me our Sogard overall and Taylor economy. So brief overview of our projects the problem is that we are relocating one and a half million books from Georgia Tech's library as well as ingesting a total of two and a half million books to the library service center and that figure includes Emery's collection that will be housed there our methodology is that we first define the problem as a team we are somewhat unaware with how library service centers operate what a library service on or even is so once we figure that out we identified the data needed to be collected and went and collected it and we also validated that data now is time for the interpretation of that data in our final deliverables to the client you will be the library relocation portfolio as well as a cost function model and this relocation portfolio will be kind of like a start to finish just how everything goes whereas the cost option model will be predictive of cost based on what types of transportation you may use what types of book truck or car you may use etc. The client obviously I'm pretty sure that that one's pretty straightforward as sort of tech librarian Emory library the location for Georgia Tech's collections are on campus in the library of course and a warehouse off site for Emery's library they have one on Clifton Road and they are also housing books the heinies evil art room forward the problem the Elysee is currently under construction and won't be completed until October November of this year also the library is looking to renovate its floors for student use study space etc and they're looking to start that project April April of next year therefore this gives us about a five month timeline with which to complete this project of moving these want to half a million bucks and just in this one a half million Additionally there is a twenty four hour cycle time the books are in the library so that students and faculty can use them in. And while they are being transported or ingested people can't call them so we want to minimize the amount of time that the books are not called Will we have been given if you are per book estimation to move in just clean process everything or book rights and that's going to be injected into the Elysee. So moving for a brief overview of the methodology. We had some LS site visits those were at Duke University University of South Carolina as well as University of Georgia we did research on what Ella sees are how they operate just kind of day to day operations versus an initial in just what the differences are. Just too and we also built a process map and this process map is start to finish just kind of in the eyes of a book so if it's on a shelf in the Georgia Tech library you have to pull it from the shelf it's going to get process transportation finally into its resting place Hello Nancy and as far as data question invalidation we had to determine the data we needed in this case a lot from the process map the process of all the showing in a in the next slide and it just breaks down everything into easily understandable chunks and so that data that's needed for that chunk can be found we collected the data needed and we verified the data it was statistically significant and now is the forty interpretation part we are building a simulation model using Samia and we are also using a function model that is built inside of X.L. these two things will work kind of hand in hand and what I mean by that is that this in your model will give us a better estimation of the time that it will take to complete the project whereas the cost launching model will give kind of an approximation as close as approximation may be for what the total cost of the project is going to be. So for expanding upon our library service center visits we recognize that many of the L A C's use the same type of equipment and you'll see a couple photos on top you'll see a photo. The Raymond Picker at U.S.C. and the backing table at Duke and we noticed that these equipment he says were pretty standard as far as Alice C. Use an operation goes. Another thing that we saw at multiple different locations were the Griffin book trucks and just so everyone some of their big metal with the large shelves they have a lot of capacity and so those were used for both in just in day to day operations at several locations and that book trucks and plastic totes were used for initially transporting the books from wherever they were located to the Elysees the processing where I was very a lot for example due process is a very high volume students are calling the books in their collection quite often their processing space is relatively small when you look at US C. It's kind of opposite in that they had a large processing area but their day to day operations were nothing in volume compares compared to do. This is the overall process map to which I was referring in blue you'll see all of Georgia Tech's collection in yellow you'll see Emrys collection and when they converge at the Elysee it's represented in this green. Emery's books are being transported Georgia Tech books we have to figure out a way to get them from the shelves to the L.S.E. we also have to decide if we're doing processing here or there cetera and then finally they have to be scanned into the system and trade in the warehouse. So Taylor will discuss more you so in this last couple slide that we have all of the processes that are involved in the overall process broken down into departments or stations so at each station there is going to be time necessary to complete the work at that station and also there's going to be equipment that's going to be necessary to carry out the job at that station as well so to get a feel for both of those figures we're going to do research online we're going to find industry standard. As for say how long it takes for a grown man to walk one hundred feet these are all known we don't have to run our own experiments to get figures for those those are readily available and as well quick costs we can look up pricing for any equipment that we might want to use on a solution or an alternative solution but for the things that we can't find online like how long does it take to size a book over or how long does it take to write an elevator we might actually have to run some of those ourselves so since the time studies are already broken down by task we're just going to record it after we reenact each one of those tasks ourselves record how much time it takes and ensure physical significance by taking a minimum of thirty samples for each task. So for the book sizing example each one of us actually was busy in the library we would take a stack of ten books we would size each of them on a sizing template and then place it down in that and another track so each each stack was ten books so we would record how much time that each that it took to process all ten books and then we could divide by tend to get just a book value of her book value and then after we do thirty trials of these and we're changing up between these trials we're changing up who it is that's actually doing the work because everyone works so to get a better feel for an average work rate all of us participated and also the the thirty Char I'm sorry the. We use different sized books also for each trial to simulate the the mixture that is existing in the library. So once we have all thirty of our samples we can actually use this expert software and expert suggests distributions to use on data streams that you give it so with the thirty points that we have on book sizing it suggests a mean and distribution that we can use to. Actually capture that work rate inside our simulations and it's really it's really handy that expert fit outputs to the exact same simulation software that we're using which is to me. So in Sydney. Looks a lot like our process overall process map the each of the stations you can see they're all tethered to one another by by paths so as one station completes work it can send it to the next station and then that station has its own unique processing time it's one of these processing times are editable and it can be directly imported from the experts that software that we've got so we're using the realistic distribution that we've got in our experiments. And the cost function model is going to be running in tandem with the simulations for each solution that we have an option for there is a unique equipment set for that specific solution and also a labor set for that specific solution depending on what it entails so the labor costs are all going to be calculated in terms of labor hours by task and then all of those can be summed together and multiplied by a worker's wage to get an average or to get an estimate for the amount of labor costs that you're going to experience and also the total function the total cost function includes the equipment costs that are going to be necessary at each one of those stations as well. So the simulation is going to give us an overall time of completion for every single task as they interact with one another and the cost function model is going to help us first determine an initial balance for the workload we're going to get work rates for each one of the stations and once we get an idea for the rate at which books are processed at each station who each station we can balance the workload so. So if the books are sized five times faster than their vacuum then we'll probably put five extra people on vacuuming just so their work rates are balanced but in Simeon any time that there is imbalance there is either going to be a starvation of work at some station downstream or there's going to be a build up if you're not performing fast enough for your queue is going to grow and we don't want that either so all of those will be made obvious in the simulation. And I'm going to pass it on to suffer for the process later. So as you can see here this is just an example of what a process layout can look like this is worth all the processing is done at the embassy as you see we try to determine how many stations we need and here we have four stations per line of vacuum station size and train station and this will scan station a verification. Station. With this we will be able to determine the total speed of a line and from here we can determine how many lines we need based on the total time so this is just an example with three lines another thing that will take into account when determining the total number of lines needed for the collection is if the which is personal for two you can delay so we will take that two in consideration as well when we figure out how many lines we need. So once again to liberals is to have a book relocation profile a portfolio start and this includes the equipment labor commissions more so what type of equipment to get all of the the options of the different equipment and the amount of different Quitman and the amount of labor per station and process as well as processing layout configured configuration will be processing layouts for it add Georgia Tech and or L. A C. depending on where we process and also for the ingestion and the daily operations as well as estimated time as I know that the Georgia Tech library in the High Museum has a very set life so we want to make sure that we meet this deadline. This timeline or minimize that timeline as much as possible for the cost function model we will be breaking down the model based on processes and give you the cost per each process how many labors you need and with a timeline of the library in the museum you'll be able to use Put your inputs and there on once you receive the model and it will relay it would get put a different cost a penny on the time so for instance if you want to work seven days a week or six days a week it would give you different cost based off that time another thing would be to outline the fixed and variable cost so there will be some variable costs such as the number of stations you have in. And the line and some fixed cost a box truck. And then we'll also give you guys the estimate of also show the Essman labor hours per process. So the value is to minimize that two dollars a quote that you gave us at the beginning in semester and doing that by streamlining to minimize that while streamline the process for the doors to collaborate High Museum collection to be within that forty five month deadline or less another thing that we would do is determine where the next are within the processes So for example if the elevator is a bottleneck we want to make sure that we can make the process more efficient by maybe adding another panel to the elevator to make it where it's faster faster flow and that's just an example of how we could add value is determined those bottlenecks are within the process you know the thing is you guys are very big on the knowledge sharing so with the data that we collect in the proposal that we give you guys. I will see our universities around the nation will love to have that data to figure out what's the best way of treating their own ingestion into the sea. So our next steps will be to continue and validate data to. Finalise our simulation model and to further enhance and build our cost model and within the data collection we will be collecting and researching data and costs and putting those data into expert fit to get the distributions to put them into our simulations then we can get the overall time it takes to complete a certain process with our cost model we will be doing some research on cost as well and putting those into our model to determine what's alternatives or with solutions the best are appropriate for you guys. So thank you very much for listening to our presentation and and the questions. Well. So. So one thing that was suggested to us is you could add a panel to the elevator and which. Basically you have control over the elevator to go to specific floors. And so won't be usable by a specific elevator won't be usable by that's my students at that time. And right now we're operating if actually bringing. Back the project we're also going to hear that different I mean. I just can't really call all the time if I think that dockery. So. There. We. Yes The question was Is there still data that you need to collect for this project. That you haven't been able to collect. There is some data that we are still trying to. Research or were able to ask other L.S.C. such as how long it takes to. Put a tray onto a shelf Adela C M I something that we can't really say model as a time study as ourselves so there is some data such as that. Trying to ask other Eliseus and see what their opinion is not based off that. Here but. Prices at times followed. You know. We haven't seen verified ourselves only because if we ask those other we get maybe a singular data point and we're working more with distributions and variability than we are we just basically means I'm so it is very helpful to have those means from other L.S.C.. Well if someone is much better than the other then what are they doing differently but as far as for this project we have been coming to the times and ourselves just so that we can have a distribution. The question was what were the biggest insights from our trip to the different Elysees one of the biggest insights was when we went to Duke University they had a huge number of the Gryphon carts the steel book trucks which are the ones that fit directly onto the forklift pickers that they're using so at first we were considering sending the books over by carts that the library was going to provide for us but considering how many that they were using it became more viable to perhaps buy like a larger amount of those at the beginning because of their capacity one of the smaller library carts only carries about like ninety books and one of the Gryphon carts fully loaded her will hold almost two hundred so so it would definitely cut the at least like the fetching and transportation process almost in half just by going with that simple upgrade. It's right there I believe but I just. Like. So one of the things that we learned was that a lot of this size or this and. I know. This process. Well I think I think I mark that this is something that we had brought in for the last race the prior year and we noticed that all the air. Visited all bar codes were and we were in the same corner so that when they were looking through a tray they were just having to like scan each one instead of having it open I was going to be inviting stand a barcode on the inside. So we built into that for storytelling Amery just because we noticed how much time it without having to open all those books but we did notice a few key differences between our allies and the other Eliseus one of them is that the other allies are extremely active they're going in and out so. You know you are retired is going to matter alive whereas what we've understood with Emory and or tax collection this might going to be quite as our news crews are going to be calling the books quite as often so it's not going to be as impactful Additionally the software system that our analysts use going live as opposed to the others they've all use G.S.A. but reason deputy fire going to be mired as it allows us to also see the title so I also cut down the time because if you go to a trade looking for a book you don't have to look for the bar you can just look for the title. Area that we would out a lot of time would be just scanning I was bar post See that's the right book you're looking for but you can do it by title but it's. Really separate. And. More. Than your question. So. If you're still a very bright bright Only five year olds. And the question here is to this day still and we can make it we can accomplish that. I guess I haven't seen any data that tell us a completely impossible or ridiculous. And we have been doing some preliminary price comparisons and can't say for sure that there are concrete concrete I all but just with what we've been coming back I think it's completely reasonable costs. From. All his friends in the real world yes. Or. Good question question was that in the real world would you ever tell a lie about their project is not doable and our advisor in the real world of course for us to suggest that you tell management what it will. Be answer it was it can be done it's done however it will come yes or it will cause that So here is a better solution. So you never say no. You say. Or at least say. You were one hundred percent right one could say. That some are about. As bad. Really. WORK FOR YOU ARE say. They are. Yes. That's it we early on we received CAD files for the the facility that has yet to be built but we already know we already know past that was one of the things about doing L.S.E. visits to existing Eliseus was that we got an idea for what their ingestion rates were and about how big their processing floors were so the LS see that we're making them Tek L.S.C. has a five thousand square foot processing area which is nearly like four times as big as and how they're a place that had a processing for so it should be it should be all right for the volume and we've also looked at potential for layouts how we want to lay out the vacuum machines and and workstations inside the processing machine before it actually gets picked up by Raymond and put on the shelf. And we also have we also have detail information about the plan of graph about the way that the shelves will be assembled inside so we know you know depending on the mixture of books we know how much that Raymond has to travel to go to each and every corner of of the climate controlled. Stocks. So you know. It's much larger. Stores Yes. Ours is our system larger. Currently it they're only under construction for one module and that one module will hold up to four million bucks but they plan on expanding it and to have another module of equal size. If you're. A storage. And right space you have to learn to love Duke University the only. Rials group of zero zero zero zero zero smallpox those who know you so it does vary a bit on the size of the module. Or a story with universes of. Your story your prophecy pasty are just barely. Because it's based on your ins it out on the sun. Earth. For the story. To just. See. Here. Your eyes. Here you know. Very well let's try. But it is a very different. Story here. Now. Up. To the very high point we only have to interject something. That's. Probably a little. They're very hard right. So they're. Less likely to be correct. But is there any clarification that might help. A. Little. I sure am. Usually because we are focused so much on the ingestion process or my summer day we are still focused on a day to day operations but volume wise comparatively it's just nothing to be happening and I won't actually be adjusting so as for right now I would say that. We. Are considering right. Rate on. You it. Was during you did want to check yes we had the mezzanine or the the I think that's what it's called the mezzanine area was not included in the processing for space. Station floor so that's really what it. Looked like Boston area. OK. Yeah we would have to I have to check the cap file just to make sure. Right right right just wait your calculations OK. I have a question for every one hundred times a day just to these two really really nice if you're giving us your variables in different markets for special. Different types of errors in handling and if you're in your city be a different rate coming out for all you know question she says and speak to the solicitor who need solutions or. We are moving the questions systematically so. Right now we're considering removing the question from Georgia Tech library for four and a lot of our special collections are in the basement so I understand the processing of special questions is going to be different than general collections right now we're considering. Right now one of the options is this and then through the regular processing line just have the time very little bit right but we are also considering having a single line dedicated to special collections because we know that the processing is somewhat different. But the special collection there is of these an amount of O.E.M. but compared to the box the microphone the microfiche it's not a huge amount but we are considering in the process just the different processing methods in the two thousand times may vary with anything special collection. You. Do all your skill. Worse. Which is the. Way. Forward. In the. Worst. Disasters are. We. Must do your. Best. To read. Our file deliverable would be the last week of April and that's when we'll be. Really. Our goal is the last week of April because we will be going to the real world and graduating. Final presentation are you. Going to guess that they will actually enjoy that presentation much like yes. We'll I think that you know more. Because we look. Like years from a few. THANK YOU THANK YOU THANK YOU. Thank you thank you.