IoT Leaders
IoT Leaders

Episode · 3 months ago

Maximising Asset Value: The Importance of Data & How to Use it Properly

ABOUT THIS EPISODE

Many companies are under-utilizing their data—a misstep caused by misunderstanding the value of the asset, how to properly mine the information, or how to blend the tacit knowledge w/ the explicit data.

Dr. Satyam Priyadarshy, Technology Fellow and Chief Data Scientist at Halliburton, joins the show to discuss his opinions and strategies regarding data and how companies can take advantage of the bulk of their assets.

What we talked about:

  • Making Sense & Insight Around Data
  • The Importance of Historical Data
  • Strategies for Mining Data within a Company
  • The Application of Artificial Intelligence
  • Hiring Different Types of People at Halliburton

To hear more interviews like this one, subscribe to IoT Leaders on Apple Podcasts, Spotify, or your preferred podcast platform.

You are listening to iot leaders, apodcast from si that shares real iot stories from the field about digitaltransformation, swings and misses lessons learned and innovationstrategies that work in each episode. You'll hear our conversations with topdigitization leaders on how i ate is changing the world for the better. Letiot leaders be your guide to iotete digital transformation and innovation.Let's get into the show. Welcome to the latest episode of io teleaders, the podcast that aims to demystify be complex, intriguing, aworld of it. My name is nice, your host and i'm the ceo of si global io tcompany and today i'm delighted to have my guest on o t leaders, dor, satiam,prettsy and satan works for a halliburton, and his title is a chiefdata scientist and he's also a technology fellow at halliburton, andif that's not enough, he's involved in several start ups and is a seniorfellow at judge, mason university on side of security, as well as a nagenprofessor at at georgetown university. So definitely a busy man and such avery, very welcome ad to itle podcast. Thank you. Nick thanks for the invitethirteen, i yes or a little bit addition to the profile, i'm not no longer with jurdon, but imoved on and i'm now with virginia tacker and oklahoma state and inuniversity in india. So is a lie! What on to it as a formeracademics, you never leave academics, you see yeah yeah, it's well! I don'tknow how you find time for all it, but hank you getting him time for this halfan hour so that we've probably been being together today. So such a it's abig subject- and i know when we spoke pride to this- it's a pretty wide,ranging subject as well. Also, let's try and break it down into pieces.Maybe i can just ask you perhaps just a little bit about yourown background before we dive in how what was your journey like? How did youdidn't you get to where you are today a being create big in scientists forhallie and perhaps for those people who don't know how the bet, because mostpeople do e the will people who done maybe just a little bit of view of whathale burton does in the on gas vang? So, first of all, he let us talk abouthaly button, then i'll talk about my journey. Halliburton is hundred and twoyears old company, it's one of the world's largest energy services company.I think one of the first pattern the company filed was in areas of cementingby mr haliburton and that's the company named after him, and it is over theover the decades and a hundred years. It has gone through a lot of expansionin different fields, but primary late is a what do you call his company,which actually collaborates and engineers solutions to maximize theasset value? No, it's a very, very important part to remember it's aboutmaximizing the asset value now, in order and days, we would just call thehydro carbon as an assent, but in today's world data becomes anotherasset. Okay, so i just work in the ground in the oito that the data is inplebeian, which is a o. We're gonna go right. Absolutely so, and i said thecompany is global. We have for about fifty thousand people around the world,pretty much presenting most nationalities in the world and, ofcourse, the challenging task of billing completion exploration, all atour olever complex scientific and...

...engineering waste. So a lot of highlyskilled people are in the company as well. So it's a great place in thatsense, because you get to interact with mathematics, engineer, scientistphysici go physicist. If you look at this spectrum of talent, that is therein the company's significant and for my own journey. So i give my phd inconter mechanics applied to biophysics when i was pretty young and and pretty concept of how to. If you think in avery simple terms, you know solar self. We make an efficiency of fifteenpercent to twenty five percent, but nature makes all ourselves with almosthundred percent efficiency. So my phd was like trying to understand what isgoing on inside a chloroplast from a quite a mechanical point of view, but ithink i was way to ahead of the girl because it just can't compute anythingbecause it involved a completation of five five thousand by your fivethousand metrics and an enornous part metrics. You can compute easily becausethere are no super computers that any went today announced pasmer c thechallenge and then i switched on and i was trained by a adviser. I think i wowmy success to him a lot because he trained me in such a way that you thinkof the problem, not as single problem. It's a multiple problem issue and youshould always be open to interesting, multiple problems at the same time andas a result, in fact doing my phd, he said not to do work on. He won't giveme this phd. Unless i publish more than topic, you know so i is is really goodbecause it opens up your mind to different problems, and so i went andswitched career, not carriers i went to post do play, did to first docks inaustralia in a totally different field like glassy dynamics and lipidmembranes, and then i came to us and it work in dna, electron transfer and nonmedia optics and many other topics. So we have a very special model, calledbarton and pratesi model for d electron transfer. You know an the. I did a lotof super computing, but if you think a foundation is all in daytail beingcantamanti ut genya to be dame- and here i was the generator of the data aswell as analyze analyzing their data for time very complex problems, and soi switch my careers from there to become a technology static companycarleol, which is america online, that botines and lucky. I was lucky there that sevenyears i went from a individual contributor to becoming ahead ofresearch in two thousand and five two thousand and set up is and profexcellence in two thousand and five in beijing and danger. And then i did my mba during that timeand switch my carriers again to become an executive turn around some companiesat involved in startups and as a resell. One day i got a call from ali mutton ora recruiter from ellen gas in rusty, said they're.Looking for someone like this, and with this experience i am having fun forlast seven years in ally burton when i set up my center of excellence for abig data, dita science in digito in two thousand and fourteen when nobody wasthinking about getting the industry. Well, there's enough material everabout twenty podcasts from us to halliburton and big data. So so,let's, let's just pick that last subject and go to deed, because you said they're instint. He said thatthe p nobody was thinking about it a while ago in the sefenteen, of course alot of people to say. Well, i actually the pride for my team is the datealthough having it would seem to me that in certainly in the field thatyou're in there's almost there's no shortage ofdata, there may well be a shortage given insight as to what it actuallymeans, but it seems like the amount of...

...data that without technically able tocreate it, i mean it's not a driving time or even running to pap it it's afit of the host pipe it. It appears to be almost like a sunam, and it's onlygot in to get more and more and more. I guess your work must rotate around that in terms of howan ursly makes sense and insight and interbase this data that is coming upas in the context of the business value of the data. It's absolutely so. Youknow people can challenge the al. My comment that people are not thinkingabout it, but that's not. That is to an extent to but the industry has beengenerating data, no doubt about it right and but if you look talk to youexpert, they will say that or even today. This is we're talking about othousand and twenty one people of the saoaies of not good quality. It is notcomplete. So if you think of where the world is, and if you look at datanative companies, which he well was one of them and now the googles of theworld, you can call them now if they also had the same problem of data, notcomplete and have bad quality than they would not be making money right becausethe money comes survive, analyzing the data building recommendation in inwhatever it is no. If, in an engineering form, we keep saying thedata is of not good quality, then what have we done to fix that right andthat's where i think we come back to more details later? I think that i utsan industrial iot will really make sense, because that can actuallyimprove the quality of the data right now. Industry collected the data and nodoubt a significant amount from exploration to drilling and completion,and i just see and other areas. But if you look at it, the data was collectedin the field. It was transmitted to the back office at a certain frequency. Sonot all the data was ever moved to back off this, but if it was moved, it wasmoved in a later state, so osa. So there was no real time or to sayanalysis of it to the extent that it could have been done now. So if youlook at even the last twenty years, evolution of the technology, forexample, surch technology, like we all depend on it right for allour practical life. We depend on it, but when it caps to implementation inthe energy sector, that has been really poor. But if you go to any of the bigenergy companies and you want to search for him, say irrotational, you have used forlast ten years and you have repaired it twenty times and if you want to searchfor that repair long, you will not find it. Why why we hear these stories a lotabout? It's almost a paradox in a way thecompanies that arguably could benefit it. The most fomething o like t thisdevice has broke up in twenty times. On the run, it's a broken for the twentyfirst time. Can i just find out what to do, andthat would save me a lot of money and there's so much money involved and oilexploration or example, is i massively capital intensity? It was a veryexpensive business. It's all about collapsing the time you hear thesestories and then you also here a cool not just a bit, but large companies ingeneral often are not the ones who have catching the learnings captured theknowledge and they are just some extent behind. U, why is that is? Is it acultural thing? Is it is it do you believe? Is it a volume of data thing?Is it that they were just pouing on something else as a priority likefinding the the oilings? What do you think that is so? I think my my take onthat is that you know whiles technology world. The industrydid not spend any in like. I did not spend well on actually understandingthis technology well right. So whoever inventor came and said, oh here is mysearch platform. You can implement it...

...to search your corporate website andthat's so got implemented. No, nobody looked at the use case. How it could bereally will be done for other areas or conto of a data mas right, so those twowere not caught about very well that how how the value will come and alsoone of the t, it's a kind of a cultural thing and callquality business strategy when you are in an operational mode. Your goal is tofix the problem and move on yes. But if you look at the whole framework of datascience is that you want to do the science from the data. That means you,your historical data is your really good source to do that. Science:experiment right, because you already broke your pam pored, bro, yea yea. Youhave to have the day now i can actually analyze that dataand understand that for last five times, why did it break so that i can avoidthose situations? Are those conditions or can address those conditions well inadvance before it happens the next time? So the thought process has to be there,that historical data has significant value and that value could be measured.And in fact, when we did some experiments, i would call themexperiments in early twenty fourteen. We showed that how much value it isthere in terms of monetary value that that the organizations are the comindustry has lost an the city o that it yeah they're sitting on an asset, ahuge asset which is historical data, but in general, people are not lookingin that direction for the value that they're looking in the other direction,with all the new things, the shining new toy, what they're doing next newinfitting, but actually want of value in the legacy data. That's just sittingaround the spot and i'm sure in your interaction. You must have come topeople asking the question who owns the data? Yes like now, if you think of asa business, and you being a see you so if infenso for the company owns thedata yeah right. So why is this question about who owns the data withina company? So when you start thinking that the data is owned by the company,then you start analyzing it and that miss people should start collaborating.In fact, i wrote an article two thousand and fifteen. I think all datademocratization, and initially there was a pushback thing. We can't talkabout these things, but democratization doesn't mean to give away. How can we look at the data rightyou're in you're in a work flow? You have first three steps month of data.Somebody has another th. Next three steps worth of data. When you look atthe whole workflow holistically, then only you can create van when you aredoing things in asilo. You really can't create value too much. I it's excuse me such a no. I mean this value. I think knowlege about my mind, isthatwas gold under our off beat. We just can't get hold with it, but thereis there's this value in the day to this huge amounts of data it turned by the the company, but but there it's notlike there's any issues in a tisane. There is already there data. So wheredo you start? It seems like what you're addressing your role and holly got it'sso a very big problem, but i put in a very important problem in the veryvaluable pronto. So, but it's also a very broad, probably i mean there'sdata everywhere i mean it's it's everywhere and i and we're creating itand every increasing value. So what advice would you me on otherpeople who has to listening to this thing? You well, you n, i com is in thesame situation and we got all this day to n. We got all this new data withoutgetting all this need to technology and sensors. How do you go about getting astrike into mind? The balu, so you see...

...what i have been doing in for lastseven years old, say from that practice, and i've been done this for otherindustry before so. I know that it works you see, use one is you have as aco or as a people we under see you people already know where the seriousproblems are a e ass at a taich knowledge, we may not have aquantification of it precisely because it's a they've got a good idea exactlyotherwise they won't be in that seat e h. So you look at the experts saying: okayin the last five years, six years, o what thing could have been reallyimproved, and then you break down the probleminto him. What you call these prints as we call you and you want to run thatmarathan, but you want to break down in sprints and say: okay, i will take.Let's assume, take this example that i started with a arrott pump right. Ifyou were the issue of deployed that on for twenty times, and it has broken oura a a a enough tasek knowledge perspective. It has broken after everythree months of diplomate yeah us some idea that it breaks after three months now. I want to really narrow down thatproblem. Why does it happen? So you've come up with them. You come up with a solution in a waythat i can look at the data close to the three month period, five daysbefore our end as before a twenty days before and see what happened really inthat operation. So you basic what you do. Is you take a business problemwhich is you can say i will say fifty million dollars, let's say, and thenyou say you break it down the problem into chunks and saying i'm went to lookat a problem which will which i can take a date or six months, and if i cangenerate two percent of that savings, then it's a problem worth solving. Sothe concept that i say is that look at the data for aproof of value, prod, not proof of concept concept, is well known. Thatdata works. The data science works right and the data that you haveactually how much value does it have, and then you start integrating datafrom many places saying, for example, if it is a palm you can connect whetherdata you can connect if possible each or data right or connect your chemicalsdata. You can connect your deployment repair log parts logwhatever it is, you start collecting different data sets thing or wheneverwe replace a bearing of this model, it fails. So you have to start askingquestions from the data and then adding on more data sets. So you do think isco best, a eson w very yeah. Some first phase should be less than six like lessthan around sixteen weeks or less on a very small amount of data where youshow yes, the data has some value. Then you had more data than you do. Anothersmall project can proof of value. Then you say i need these five more latersets to connect to it. They could be under different silos and then youscale the problem so by the third step. You already know how much you valuehave went to generate either in costumings or leven ue generation or acuracy or efficiency or np. So it is anything you do with data has alwaysvalue. So, where you know you may have heard people saying on my digitalproject fail on my data is project fail. I don't believein that at all, because no projects fail because every project has a well.You could not scale a duct, a different issue. Well, actually, some of the danger that my company, i said that i rule- and ireally my company- knows that actually the day to the data on the failure ratepriority projects is appalling. It's some of the industry. Annalists havetalked about eighty percent of it projects never to make it pass thepc and when you double click on that the profit consent. When you doubleclick on that, it's actually nott, really a...

...technical problem there. Oh there areissues, definitely to do with the device. Most people don't know anythingabout a are design. A d don't want to know anything about, had, will designand that's a gap that we that we felt working with the modulantur instance atthe quick tells of the world and jamais the world of bols. So it fails becausethey suddenly, like turn your tap on. They stones, stare, collecting a lot ofdata and that's the point in which they just breathe. I mean they just theycan't measure the quality of what they're getting they don't know whichtines in important which they hersman they haven't, got an architect ship orwhat are they process at the edge? What data do they back or if they send itall back to head office? Think they only magine you in oils field, on a ragor something the amercan dat and taravo petabytes, and we are thy mass. Youjust can't forge to send it back to your corporate headquarters to crunchit. So you've got to do an age presidency and they just they justdidn't, think about problems of a data, architecture and big date to an insightwhen they started their their project. And so you do see what you're saying in thegeneral statistics on the industry in the people. They don't start off withtrying to drive it down to, let's just try and find out. Won't i take it aproblem. I go chase down that one particular problem after when theystart up with a horizontal approach and say: let's collect date, i grom as manythings as possible and then we'll work out what to do with our data and that'swhere often they just braced, and they think i no. I don't know i datingdanger, but i don't know i don't know what i'm going to do. Yeah absolutely. You know, but myphilosophy has been very different. You know we we can address as big as of aproblem right in today's today's environment, when a iodise, sors or anydevice that is generating the data data can be in any format. Today's worldcome because computer is so cheap, a right and an in principle. You couldreally put out very what i should call powerful machines at the edge as wellin a very small factor, so you don't have to really back hall everythinghere like whether it's cloud or ed or whatever you want to call it pognon,and you can do that right. You can develop the algorithms or the models,whether it's scientific or augmented models, and you can push them back tothe field where i, where the challenge comes, is that we we reallycan't make automated yet because we have to really test anything we do inthe field and any complex industry. You really have to test and valuate moreoften than anything, it's right. The models and models after all and morevalidation is done, but the size of the data is not should not be a concern,because if that was a concern, then, as i said, intonate companies will notexist because the problem is in sole page is exterites. Absolutely it's,it's not an inhibitor just because they want to, especially with cite theamount of processing how that's available. It is not a fit. It's not ablockage in the process is what you're saying absolutely it's more aboutbusiness problem that we want to solve and whether i need to really look atthe data on one month or should i just only look at a one week because,depending on the work lowe are talking, we could be really creating more valuewithin one week's period of worth of data and should be good enough for uslight or something we might have to look beyond it. So most of the thingswe want to eventually do in real time, and so that means euro don't reallyhave to process, but a bit some day to all the time, because no we are not. Weare not any of this. Industry may be serefiny ras. If they were to deployforty send censers, maybe they will get any tin terabytes because no, it censorsense data in gigabits, yes or day. I...

...notice i istic that the anegada, thesister or probably speaking, to the person who does an oil ring. I heard- and it was a fewyears ago- and i oil really ole excuse me- an all refinery couldhave ten million senses and i've got a feeling that a about a four or five yelartistic it is that on there or to multibeam been you know,these are numbers written by a lot of people, i'm not sure sure i've neverseen them before, like i've not been to a field, so i can't say that for sure,but what how do they calculate and what does a device means? What does thesensor mean? Are the really internet of things devices or their any deviseslike so it by definition, anything that is connected becomes an internet of athing or in in right, but if you are, if you have pumps that are running andgenerating data which is collected by hand that that's not the internet of athing, but it is a date yeah right. So these when people write these kind ofarticles, i don't know how many of them are really counted. What is connectedand what is not connected almost erle. Nobody, i mean worry the ig business and we arecustomers in the oil on guest and let me tell you to so true it vices asopposed to something with a controller and it was able to spit out later whichisn't iota, but it says a true idea places: it's probably the hundreds inpractical terms. Today i mean it's no one, i admit be in the future, it maybewith five g when we start going predate finding networks to in these locutions.But even then, i think the the word of the phrase iety is being stretched tocover everything a it fine and that's not what we're talking about in fact iscounty productive. To think that that's what we're talking about absolutely andi think again, it's a future of the web and everything becomes digital. Maybethat is when we will get that kind of censors the count of censors andconnectivity. But we are no we're close to that right and if you look at alland gas industry they're talking about digital il fis twenty five years ago,or something and integrated reservin management some thirty years ago. Sowhat is integrated of and what is it, which digital oil feel really existright, because foundation is when you say it to a feel if everything isconnected- and you are re you're doing real time- automation that then itbecomes nearly valued but pieces of the puzzle are automated, no doubt about it,but we don't have a fully holistic, automated digital oil fields, but wellanother thing people said not that allows us transition to s. One of thefinal big subjects was getae. You took about thirsty years ago twentyyears ago, but let's just say five years ago, people would say: well you know the aster to this isnothing to do with human. As you were saying, people would say by now or inthe future. Shortly it's all going to be artificial intelligence. It's goingto be machine, ran the machines will say, go the humans. Will let go he'llstand back when we were worrying about what are we going to do the jobs,because it's all going to be a brother? You know the the machines are going to beat the humans at the analysis. Theycan learn about. The public breaking they can learn about the resolution they can go. They can give you fromreactive to proactive, pre empty. I mean that's what we they see inpeople's cars, they see it, you know it would break, and so you take itsomewhere to be fixed and then it would be. That would be reactive andproactive. I the like o, we go on saying it's going to break like he'sgoing to oil, so put oil in before i bridge and pre emptive, the tester youget in the car of the room and again it's like the ice. You know, while youwere a sleep supper of date, downloaded...

...we fixed a whole bunch of two issues.You know even you you have had, but don't worry you'll never have themanyway, have a nice staying. So by now we were going to be in this wold whereor at least entering into this world, of a machines and taking over and now iknow, obwe spoke previously. You and i can see, is on dean. I know where we spoke, you you have. You have yourdoubts, whether the not the world of a fac, but you insist me that you didn'teven like the phrase o ai because of the your experience the expand on thatit yeah so so artificial in tellus. My itself, asa subject has been there for fifty plus years h and if you look at even theapplications of algorithms that are developed has been used by oil and gas in vestryfor last forty five years, whether it's near and net work, whether it isregression a right, it doesn't matter which algorithm you are talking about. The world change on the technology sideand the compute side. An artificial intelligence is just asubject right. It's like my analogy that i havealways explained. We never say that we are eating chemistry or we are wearingchemistry right where all clothes are made out of chemicals. Food is made out of chemicals e somechemistry going on right, application of chemistry that we'retalking about. So it is in the same way it's an application of artificialintelligence, whether it's related to adou weex elated, to video otherselated to data, while is related to a text text right. That is what you'retalking. So there is nothing there's no box called artificial intelligence. Ipaying is not in and when these articles come out, saying artificialintelligence in tesla is that the same box? Can i put it on my computer for for doing my statistical analysis?No right. So it is not a thing and that's where the confusion is, but inrespective of that of creates a field and it's important field, and it allowsyou to analyze things that human beings by themselves could not do all right atscale. Repeated tasks that can be optimized and even can eventually wentself learning. Algorithm will be there more mature than maybe things canactually improve like you can see those examples and robotics little bitwhether robot can learn and things like that. But when a be, we are far awayfrom it in terms of application, maybe maybe in some difference actor thinksthat we don't know. Who knows what going on there, but in a practical in apractical world? That's nowhere close to it because it requires it requiresinfici or all the tasing that is sitting in your head and all the peoplewho are actually in the field right what to do when that is in people's andengineers head after twenty thirty forty fifty years of experience and isjust a stud. And if you look at the history ofknowledge management in the companies, we have never figured out a way tocapture the tastic knowledge, thus in knowledge, yeah, and that a techknowledge is what is really what algorithms will need to really make adecision so bat. To your example about the pumps we talked about asking theengineer: abo been around bridges, statable latin. I was going toinvestigate the i got from one area. What it to be that the experienceengineer would say you really want to have to look at these posts becausethey break to three months, and i is a really big issue. I that's. It comesfrom the just the experience that does it loge the chances of the computersystem, saying that to you are probably pretty pretty low and and then, whenthey do, i was den with the data. They need that tastic knowledge thatexperience the thing that we can't codify to actually interprete the data prarieactions, and so it's a company you're saying i guess it's a combination ofyou to absolutely very, very important combination, especially in the islandglass industry. Having worked in like...

...you know and consulted in almost sevenadvertise before i came here, and i can tell you in a an ass in dusty, thepeople have so much knowledge of because the processes are complex. Irespective of what section of the work or work life a sorry work flow. We talkabout in general, the energy sector. If you look at it, it's a very what youcall science and engineering driven industry, and so a lot of these peoplehave so much to a knowledge in them. That really needs to be captured andcan be taken advantage of, for example, when you are dealing right, if youthink of it, the the person can feel and say. I need to rotate this much.The privat is not to get take i'll. Go it from to do that. You have to reallylook at so many things first year to understand what is the force comingback as, and then you had to really analyze what i did in the past fewscenarios like that, so it will take some time and and that's what i'msaying that this you, but at the on the other hand,those people who are experts, they have this sense that they feel that's thethe sound of the feel the vibration and the said do this: now they don'tbelieve the data, the data of people, don't really don't have the sameknowledge as the engineer, so they have to come together and i feel that thepeople have that tacit knowledge. They can be trained with the data knowledge,and that is what i call the talent transformation process to. I guess youwill not the the way ran the other around is hard because you, you can'tget the field experience right. Yes, yeah! You can only get what the fieldpeople tell you yeah, but that, but the fear people have so much. So you reallyneed to show them a graph saying this is what happening here. What do i doand then you quatit they was question it, because this open to a s of soy, different questions. The combination of the field experienceto give you the justin knowledge and the ability to make that instinctivehuman judgment. That says this is right, or this is wrong. The vibration itfeels right doesn't feel right. We don't know where it comes from, but wecan do it and then the m, the machine playing its role, analyzing thinkingsvery, very quickly as well. Does that mean that, in your role at halliburton, do you train people, do recruitdifferent types to people for the world that that we're heading into you know? Do you look for certain typesof degrees and reminded listening to do not the field of owning gas? But youknow conversation. Certainly, i've been about don for man many years and lotsof different industries that i have worked in people saying you know: nba is a nba.Students are useless because they don't have any of the practical experience,but they, whereas course the nba schools spill.Think that they're train the future leaders and everyone's got allthe knowledge they ever need because they got an mba. You know you take thatandyou've never applied it to an oil engineer so and often like in the caseof our daughter. Our daughter went to work for several years. Then did an nbaand actually she felt at least she was much more valuable at the end of thatthan if she had done it. The other way around. So do you in alabrand get makerecommendations in terms of what types of people you employ given in thisworld that were in already and heading more into, or do you actually run intotraining courses on how to get this combination of the machine and thetacit and the human working in harmony yeah than answer all those parts in ainteresting riso? As an academic, professor, all my students i for mb-especially i tell them, there's no point doing mb after bachelors, yetright a couple of years. Self experience then do an mba and you willknow what mistakes i did or what you didn't do right. She garten the valueof nba, becomes really important. Otherwise it you like taking any othercourse. You passed it and you're done...

...right and that's my firstrecommendation to most people in terms of since my background was not a directroland gas. So i know that it's all about generating value from the data,and i think in that way most of the team that i have built in estay are allpeople from different fields of science are engineering or other areas. So i'ma phd in atomic physics, astrophysics. I have a phd chemical engineering,mathematics economics, things like that, so they can do, they can think totallydifferently, but then you pair them up with the tastic knowledge people thatsubject mate exports that helps and then over the years we actuallydeveloped our own training program not only for individual contributors, butall the way to the leadership, because one is, you have to really keep these peoplealso in house right to there's interesting challenges in the world andif especially, a data science, people their high high and demand. You know ies o jobs at there. A ring a future ranges exactly so the way myphilosophy has been that for all the data scientists give them interestingproblems, don't give them and put them in a box and do just one problem ifthey are doing multiple problems at the same time, there's no problem with that,in fact they love, because then they can think of it. A that. I have thiskind of data, how these are the algorithms working, but i have thiskind of data. Why is this not working to? They have their own com? Compareand contrast going on within themselves and and then they are interacting witha of different padomano experts or to say, and that helps them really thinkbeyond a simple problem, and then they it's an exciting environment, to workand that's how we have grown this center in bangor and houston and columbia and many otherplaces we have working with so many people, but the training part welldeveloped is because the same people who are actually working on a problem.They are actually teaching the house and house of thisfield to the domain experts. So they, when they ask questions they learn fromthe roman expert. Why are they asking this question? Why can't i find thisand when the roman experts us how to do mathematic tell y or why it is likethis, they can explain it and that this energy is significant and last, i thinklast just in last two years, i think we have played over a thousand people inthe industry and so and and hence i don't really worry about the talentpool side of it. In fact, i and one of the hat iber is the managing direct ofindia center and in the last one year, i've hired of you about hundred people from all different fields. So so youknow it's a fascinating area to work in and i think the potential issignificant. As i say, the opportunities are significant becausewe are only scrash the surface of the of the industry, and if you really havethe desire to build full implementation of iotete properly getting the five gnet work, working or beyond five, your working, which will reduce the cost tomove the data and ring the speed to the connectivity, then i think we will have to build what is calledthe digital tin of digital twins and so there's a fascinating field. And ofcourse the tacit knowledge is not going anywhere. It is. I call it augmentedand augmented analysis going on yeah it's when it's a fascinating story. It's afastning journey and it's also reflecting them so the way of halliburton's journey as as be companyinto moving more into dat and being to services, boa the clients. Of that yousay all the efficiencies and then the whole subject of digital twins issomething that we do plan together as well in a future podcast bok of the o.Would we better leave it there because...

...we covered so much so much ground, soadia, just penashe lit some people, thanking you for your time and sharingwith lorises your journey. What you're doing in your thoughts on having girl,marted and for also so strading those people who were hasse were out therebeing concerned about whether the michigan's gin take over the actuallyyou don't believe that they will e and that we're all going to plenis do goingforward in future years. So with that, i just want to say thanks for everyonefor listening, you, men listening to the ant leaders podcast with me, aperson girl, if you have any seen bat or questions on it, to remember that wedo have an email address, withis, io, t leaders at s. I think that's es ye com,so we look to give from and he suggests from any subjects that you would likeus to cover. As you know this this particular podcast, we can actually govery broad or even vertical into industry, and we love to hear from youas to what you living to a discretion about or even when you feel you'd liketo be guest on the show o so tested. There sat an thank you very much foryour for time and for our listeners will see you and talk to you on thenext exit thanks pring an or nip by not thanks for tuning in to iot leaders, apodcast brought to you by si our team delivers innovative global iot seulaconnectivity solutions that just work helping our customers deploy,differentiated experiences and disrupt their markets, learn more at si com.You've been listening to io t leaders featuring digitization leadership onthe front lines of iot. Our vision for this podcast is to be your guide to itand digital disruption, helping you to cut the right rate to success. We hope to day's lessons stories.Strategies and insight have changed your vision of biot. Let us know howwe're doing by subscribing rating reviewing and recommending us thanksfor listening until next time. I.

In-Stream Audio Search

NEW

Search across all episodes within this podcast

Episodes (18)