 Think Tech Hawaii, civil engagement lives here. Good afternoon and welcome to another episode of likeable science here on Think Tech Hawaii. This is our special likeable science series in collaboration with the University of Hawaii's College of Social Sciences and I think this is the second one in this series. With me today from the Department of Economics, I have Nori Taurui, excuse me Professor there and also associated with Yuhiro, a very economic research organization. And graduate student, Asahi Oshiro, getting ready to get her doctorate in economics, right? I just graduated this semester actually. Congratulations, wonderful, wonderful. And they're a fascinating piece of work right now. They're looking at something that, you know, likeable science is all about how people relate to science and all and we all use electricity in our lives right and we all pay our electric bills and you guys are looking at sort of how those two factors play against one another, right? That's right, yeah. And you're doing this in collaboration with the Hawaiian Electric Company because of course they've got all this data right there. The ones who gathered the data on electricity used and the bills that are paid, right? So why don't you go ahead and tell us maybe a little bit about this study, how it came about and how long it's been going on and all. Yes, yeah. As you mentioned, Ethan, so this is an outcome of like a little exciting collaboration between Hawaiian Electric Company and University of Hawaii at Manoa, in particular, Economic Research Organization, Yuhiro. And my colleague, Professor Dennis Conan, she's now the dean of the College of Social Sciences. She started the initiate collaboration efforts with Hawaiian Electric Company about like seven, eight years ago. I joined the team about three years ago and right after that, Asahi joined the team and so the way it works is that we send our graduate students to Hawaiian Electric Company and students would work with the data, the confidential proprietary data that Hawaiian Electric provides and provide data analysis services that would be useful for Hawaiian Electric Company and at the same time that would be useful for academic inputs in the form of students' dissertation research. Great, it's a great win-win collaboration. Exactly. They get extra services from people like you basically and you get rich data to mine and you do your research. I'm glad to eat. With a degree. So you've got a shocking dissertation, right? Sorry. I hope so. And so what's sort of the driving question of this work? Okay, so the basic motivation of this effort is that I think as many of the audience might know, state of Hawaii has a very ambitious clean energy goal with 100% renewable energy by 2045 and so what we see is that as more and more renewable energy is used, the ways in which the utility company operates might change and also the ways in which customers, consumers pay for electric bills might change. Okay, yeah. And what was your particular piece of this work? So what I do is that as I act as a data analyst that's sent from the University of Hawaii and I get the professional experience and you know I get a lot of hands-on experience I would say with the data and then I present an analysis that would benefit both Hawaiian Electric Company and myself as part of the dissertation process I would say I think it was a good experience because not a lot of students you know get the chance to be able to be able to work in a you know professional environment and be able to have a lot of hands-on experience. Sure, that sounds like that'd be great. Most graduate students spending their time out in the field or buried in a lab somewhere. Yeah, exactly. It's a good experience. You get to talk to the other professionals in the field as well so I think that was another benefit that I got. Exactly. Excellent, excellent. So why don't we jump on into this and start looking at what are these you found and you know. Sure, sure. We got a first image here. Okay. So in our project we look at direct-to-sale consumption in the business sector in Honolulu. So we look at various different types of sectors such as department stores and others. So what you're looking at the profile of electricity consumption and how it changes within a day from midnight to morning noon afternoon to midnight and also over seven days of a typical week in 2014 in this case. Okay, so this is department stores here. Yes. A very regular pattern we're seeing every day sort of peeking rather steeply early in the day staying plateauing during the day and dropping off in the evening. Exactly. Right, yeah, I think you made a good point about the picture. So one thing that we notice is that within a day the electricity consumption fluctuates a lot. The department stores when it's open, it uses lots of electricity for air conditioning and other purposes. But during the nighttime it's closed and doesn't use much electricity. At the same time we see that for seven days a week this fluctuation seems to be fairly consistent. Right, right, this is strikingly even graphier. Excellent, excellent. So this is with department stores and the next image. But not all business sectors are alike. Right. This is what surprised me, the hospitals. Yes, so here we see that with hospitals in Honolulu we see similar cycles and fluctuations, but within a day the extent of fluctuation is smaller. Right. And also on the weekends we see that the consumption of electricity is much lower than during the weekdays. Yeah, that surprised me that you see that dramatic difference. I guess people don't want to be in the hospital over the weekend. Hopefully not. Great, yeah, interesting. The next one. Okay, so why do we care about those fluctuations in electricity consumption? Well that's because if you look at the cost of electricity generation that changed quite a bit over a day and over a month and across seasons and years. And that's because electric companies would need to activate more expensive power plants when the demand for electricity is higher. So you're looking at the picture of the red color indicates the actual price that the business customers pay. So it's flat. Regardless of the cost of generation, typically the electricity rate is fixed. Right. But at the same time if you look at the red color, you see that hour by hour the cost of electricity generation is different. Right, right. It was the other line. Very good deal, right. Super. So that suggests that there's a sort of mismatch, right? They're paying a very constant rate that is not reflective of the actual cost that costs the electric company to produce that power, right? Exactly. Okay, okay. Should we move on to the next picture? Sure. So while you mentioned the discrepancy, but so this discrepancy is going to matter a lot as the state of Hawaii pursues its renewable energy target. And in such future with lots of electric renewable energy going in, it'll be better from the society's point of view to align the consumption pattern and the generation patterns, taking into account the ever-changing cost of electricity generation. Sure. So walk us through here. You've got some different industries you call the education, hospitals, grocery, hotels, and merchandise. And they all seem to have these two lines, one says current and one says alternate. So why one of the patterns is different and two are sometimes the blue line, alternate is on top is higher and sometimes it's lower. Okay, I would like to answer this one, especially because this was one of the outputs of my efforts at Hawaiian Electric Company. And just to briefly explain, most of my research is focused on the commercial and industrial. That's why I have the five big sectors there, you know, the education to the merchandise. And so what I did here was that, so as Professor Terri mentioned that, you know, what would the bills look like if we didn't have a flat rate structure as in the previous slide that we saw? So what I did was I calculated what the what the customers bill, what the customers bills would be under an alternative rate structures, which actually incorporates the hourly changes in the generation cost instead of facing just the flat cost, right? So that's what you see in the two lines. You see the orange line is, you know, the estimate of the current bill that they see for each sector. And then the blue line is the simulated or the calculated bill that I calculated using the fluctuation in the generation cost for the utility. So why is it that some of them, for instance, education and hospitals, the alternate scenario seems to be less expensive, I'm reading this correctly, but for some others, particularly the grocery store, is the alternate scenario is much more expensive. So there are a couple reasons to this, and this is part of my dissertation results. And I did find that whether or not a customer's bill increases or decreases under the alternative bill depends on actually their load share of the total load share by sector. So for example, I mean, to put it in more easy words to understand, I would say it's close to something like how the volume of the electricity that they use is a lot more bigger. Okay, so you're saying so education, the educational sector would, I would say would is a little bit bigger as, you know, they hold a larger share of the pie compared to the grocery stores. And actually, I do have a couple slides, I think, I think the next couple slides would actually show that share. Okay, let's go to this next. Okay, yes, here is right. Oh, so this one is just zooming in on the educational sector. Right, so and it's showing indeed that the current bill is higher. Yeah, that's, that's the right. And the alternate section, if they're really paying for power for what it cost to produce, it would be somewhat lower. So they should, the school systems and should want to go for this structure, right? Ideally, yes. Okay. Yeah, so here is one big reason why you see differences across sectors. So another reason is, for example, in the case of educational sector, they tend to use lots of electricity during the daytime when the cost of electricity is lower. So that's perhaps another reason why they tend to expect lower electric bills. Okay, okay. And then we go on to, I think, a detail on grocery stores in the next image, right? And here it's just reverse now, right? It is reverse. And this is because, you say, because grocery stores are using less overall than also using a different pattern. Yeah, exactly, different patterns. So when you look at how they're using electricity during the day, it's a little bit different than, for example, the educational sector. So maybe that's one of the reasons why you see, you know, the opposite phenomenon than the educational sector. Right, so they're not going to vote for this alternate strategy, obviously, that's going to cost them more, right? Exactly. Okay, okay. That's intriguing that there's that dramatic difference I find, particularly because both of them, well, they have some differences there. Basically, again, it's sort of their main use is during the day for both industries, right? But it does depend on the magnitude of how much they're using in the day as well. Okay, okay, interesting, interesting. Then the next one looks at some of this change in the day. So maybe, again, it's just one of the things that you do. Yeah, sure. So this one, I simulated different types of load profiles. So load profiles as in the shape of their consumption. If you were to illustrate it into a figure, for example, education, you see, so the green line is their actual consumption. So that's how their, that's what their consumption pattern looks like when visually. However, the blue line is the simulated curve that I created based on actual data. So you see that when the education sector or any sector for that matter is introduced to the alternative rate structure, you know, the one that fluctuates based on generation cost, you see that the load shape changes slightly. Right, okay. So that's the blue line. Okay. And the gray line is the generation cost that we looked at in the previous slides. Okay, that's so, yeah. And those three things and the actual, the simulated and the real-time cost producer are all somewhat different, particularly I'm surprised to see a dramatic difference between the real-time cost and the actual, exactly. So the green line is what their electricity consumption pattern looks like under the flat rate and scheduling. Yeah, interesting. We're going to dig into this more deeply when we come back. But right now, we're going to have to take a break. Norie and I saw a Oshiro from the UH Department of Economics here talking about electricity use and cost scenarios. We'll be back in one minute. Hey, hello, everybody. Thanks for joining us on Think Tech Hawaii. I'm your host, Andrew Lanning, the security guy. I host a program called Security Matters Hawaii. And I hope you'll join us on Fridays. We air at 10 a.m. And we're going to be talking about those security things that really should be important to you and maybe get behind the scenes on some of the things that you may not know about the industry or about products or even about your habits. Security is all about people, processes and products. And we hope to bring that to you in an informative and hopefully a useful way. So again, 10 a.m. on Fridays, Security Matters Hawaii on Think Tech Hawaii. Join me. Thank you. Welcome to Sister Power. I'm your host, Sharon Thomas Yarbrough, where we motivate, educate, empower and inspire all women. We are live here every other Thursday at 4 p.m. And we welcome you to join us here at Sister Power. Aloha and thank you. And you're back here on Likeable Science. I'm Ethan Allen, your host here on Think Tech Hawaii. With me today are Nori Turi and Ashiro Asahi from the UH Department of Economics, both of them. And we've been talking about electricity usage and rate structure and how each of those impacts one another. And in particular, we've been digging into some of Ashahi's work on her dissertation that looks at actual bills that people are paying, actual use patterns they're doing and the cost of producing electricity. And we're looking at a large structure, a large number of those earlier. And we want to dig a little more deeply into one of these at this point. So, and I think, yes, next figure. So explain again in a little more depth so we can really understand this. Sure. So what this graph shows you is we zoom in into the hospital or the medical sector. And the green line shows the actual consumption pattern throughout the day. So this is this graph is based off of a day's consumption. Exactly, exactly. And the blue line shows the simulated consumption pattern. So that's what the consumption pattern would look like under the alternative rate schedule. Okay. Are you saying the usage would actually be higher under? Exactly, yes. Because the generation cost is a little bit lower than the current flat rate. As we saw in the previous slides, you saw that the red horizontal line was actually a little bit higher. So that's why they're consuming a little bit more under this alternative rate schedule. Yeah. And then this rather dramatic fluctuations during the over the course of the day, an actual real-time cost that is the cost of the electric company to produce the power, right? Why does it take this dip early in the morning so strikingly? So actually, this generation cost is just for March. It's not during the year, but specifically for this month, I think you see a dip in the daytime because it could be because of solar energy that's available. So generators don't have to, you know, ramp up to meet that daytime usage. Okay. That's very, very intriguing. So compared to like hot summer months, perhaps the demand for air conditioning might be lower during the daytime. Okay. That might also explain that. Right. Then at night, you see the higher the higher in Russian cost because you're not having any solar feeding into the grid. Okay. I'm getting an understanding. Exactly. And that's when people come home from work. That's when they turn on the lights or maybe rank up the AC as well. That's why you see that. Yes. Hot water, etc., etc., okay. Laundry. That's a big one as well. Okay. So another look at one of the detailed, more detailed comparisons is with the next figure has a merchandise unit. So this is retail stores and all, right? Exactly. So the reason why we show merchandise relative to the hospitals is because, as you can see, the blue line doesn't really change as much as the green line, meaning that they don't really shift their consumption under this new rate schedule. Okay. Compared to the previous slide where you saw the hospitals, you saw that the blue line was a little bit different. Right. But here the shapes are very similar. Exactly. Meaning that they probably do business as usual, regardless of the generation cost within the day. And then again, you see the same real-time cost to production because, again, I assume you're using the same. That's intriguing. That's intriguing, the subtle differences between these. Yeah. So what Asahi is finding here is that depending on which business sector you look at, the ways in which the business responds to changing the price of electricity, where there's a tendency that you would use less electricity when it's cheaper and, excuse me, more expensive and more electricity when it's cheaper. Right. But that extent can be different if you look at a different business sector. Yeah. Okay. Okay. Because they have their own operation, you know, strategies. Sure. And there's different things. But now, obviously, sort of time of day is not sort of the only factor. Right. There's also time of year. Right. As you said, the hot summer days, people, everyone puts on air conditioning. Nice cool winter here. People sleep their windows open and let the breezes blow. Right. So I think we get into that in the next figure, if I'm not mistaken. So this one's got some detail in it. So you want to break this down for us? So actually, this is another topic that I was involved in. Okay. So, you know, what we talked about until now is more of a price response. How do different commercial and industrial customers respond to different prices within the day? But now I look at, instead of the price response, I look at the temperature response. Because, you know, with climate change, you know, there's a lot of forecast saying that the temperature might increase in a couple years. So that's why I came up with this topic and wanted to look at how different sectors actually respond. And what I found was that the green line, the merchandising, the retail sector, responds to temperature the most. You know, the striking difference that is, and it said basically when temperature goes up, the electricity demand of essentially stores, commercial establishments goes up too. Exactly. So that's what this graph shows right here. So as you go to the right of the spectrum, temperature is going higher and Y-axis shows the response of the consumption. Right. So and it's probably air conditioning is probably a big factor there because stores here leave their doors wide open. So that's one of the reasons. Okay. Whereas hotels, education sector and grocery all look much more moderate in terms of their response, right? So they're still responding to higher levels of temperatures, just not as much as the merchandising sector. And one interesting part is the purple line actually that shows the hospitals. And I found that hospitals aren't too temperature responsive. Right. Yeah. They look almost flat. Almost flat. And that makes sense because hospitals are kept purposely closed, right? Because they don't want the germs going out and they don't want more germs coming on in. And so they, unlike most other businesses here, they don't leave their doors open, right? So probably not. That's right. And at the same time, a hospital might be partly required to keep the airflow constant so that regardless of the temperature, they might need to constantly use like air venting and whatnot. That might expand this flat. Exactly. Exactly. They're always going to be running a pretty constant AC cycle and keeping their temperature pretty flat. Yeah. And another reason why you see that for the hospitals is it could be because when you look at their total share of energy usage, maybe unlike the department stores, it's not fully consistent of AC. Maybe they have other heavy equipment there that's also energy consuming. That's not responsive to the temperature. Yeah. Okay. Fascinating. Fascinating stuff. And then we have another image that's again sort of breaking this out by sector. No. Back one image, please. Yeah. Yes. Yes, that one. Right. Thank you. So tell us what's the takeaway from this? So I particularly like this graph because it shows you the consumption patterns by each sector. So this is actually something that not a lot of people are informed of. And actually I learned about this after I started my interning for Hawaiian Electric. And you see that, for example, let's use the hospitals. So if you look at the red line, you can see that they use most of their energy during the day. Right. And same with the merchandising sector, but it's not, but it's flat when they're open. So maybe they constantly have AC on during that time. So there's not much fluctuation there. And what the blue line shows is actually Oahu's average. So that's what Oahu's, as a whole, in the aggregate, that's what their electricity consumption pattern looks like. So I'm comparing both to see how they change. Okay. So that is this sector, all you're saying is from around and across the country? Oh, no, it's just Oahu. Okay. I'm not quite clear that what's the... So Oahu would include all the business sectors, including hospitals, merchandise, as well as the residential sector, like households. Okay. Then why on that multi-graph that we were just looking at, why aren't the blue lines all the same in each vote? You were saying those are averages from all the different sectors? Okay. The look might be deceiving, but indeed the blue line is supposed to be the same. The shape is exactly the same. Oh, it is. Okay. The only thing that's changing is the red line. Yeah, I see. But you're right. It's just, it's an interesting optical illusion that, yeah, but you're... I see. Okay. That makes now that... Now I see it. Yeah. And so it's very striking the difference in the different sectors do that. So this actually shows... Well, this actually can explain why I think you asked, you know, why some consumers face a lower bill or increased bill when they face a non-flat cost. And this graph actually can explain why that's the case. Because when you look at, for example, education, they're using more when it's actually cost less to provide... Exactly. Okay. Yeah. I see. I get that now. Okay. So education and merchandise, we're doing pretty much the same patterns here. Yeah. High usage during the time when it's less expensive. Exactly. Crank out the civilized... So that's one of the reasons why they could be, I guess, benefiting or they experience lower bills under an alternative rate schedule. Okay. And actually what Asahi finds about the education sector, that might be even more exacerbated into the future as the set of Hawaii integrates more and more renewable energy. Right. It's solar power. Yeah. Hopefully, solar will keep feeding more energy into the grid at essentially lower, no cost. Because with solar power, yeah, the cost of electricity generation during the day time would get even lower. Right. Yeah. Exactly. And last but not least, we have this pie chart, which looks at how much absolute power the different sectors are using, right? And I'm surprised that education is a big one. That includes K-12 and the universities. So it includes elementary to college or universities. Yeah. Wow. But this graph does not include customers who have solar panels. Okay. So just to make a fair comparison. Okay. Okay. Interesting. So, yeah. And you see the other big... I'm surprised that grocery stores and department stores are that small. Actually really small, exactly. Yeah. That's sort of counterintuitive. Hotels, I understand. Yeah. Or your condition, everything. Yes. A caveat is that not all customers have the kind of like 15-minute detailed electricity data that we use for this analysis. And the pie chart that you look at is the sample of customers where we have such data. So those are big power consumers. Okay. Well, hey, this has been incredibly educational for me. I've learned a ton. I'm sure. I hope so, too. Thank you so much both for being here. Thank you. It's been very nice. And I hope our audience will come back and join us on likeable science again next week. Until then.