Digital Transformation: Art of the Possible

thank you for joining us for our first sel interdisciplinary research center seminar for 2021. today's lecture is the first of our four-part seminar series co-hosted with o-9 solutions we are fortunate to have umesherazu executive vice president of product management and product marketing as our speaker umesh has over 20 years of industry and research experience working with large fortune companies specializing in retail consumer goods distribution and manufacturing he has a successful track record of bringing innovative enterprise applications to market in the areas of integrated business planning and business intelligence prior to o9 umesh was involved in strategic and leadership roles for enterprise platform and applications companies such as microstrategy teradata and i2 technologies umesh's research interests include supply chain analytics combinatorial optimization data visualization and network flow algorithms with that introduction i will hand the session over to umesh thank you thank you everyone thanks benoit thanks andy for the introduction uh we are extremely excited to be part of the session it's the first one of the year for you guys so we have to make it good but what i wanted to cover today is uh the topic as you know is digital transformation part of the possible so let me just switch over to my next slide here and just give you a quick introduction of myself my name is umesh and i head the product management at online solutions i have been with the company about nine plus years now but i do have a background in uh in operation research and industrial manufacturing i spent a lot of time in academia like some of you guys are doing but exciting times uh now i'm practicing in the industry so it's also exciting as well just putting the things you learn in theory into practice uh those are things that that that are really good for us so with that let me work walk through the agenda today what we plan to cover so it'll give you a quick overview of a few of the topics that i want i thought might be interesting uh first of all just introduce who's online so you have some background on what we do what is our history who are our customers then we'll talk about the key topic here around digital transformation right the the topic around what does it really mean there's a lot of buzzwords out there but executives still are not sure what does that mean for them uh and then we do have some kinds of learnings that we've observed from the pandemics of last year uh it's been a disruptive year right in terms of supply chain and business transformations so we'll talk a bit about the learnings and maybe talk about what we are seeing in terms of companies moving their business operating models from traditional to digital touch upon that a bit then in terms of current states also paint a picture of where most of the traditional companies are what is the current state um how do they move from their siloed operations and slow planning processes to be more agile then i'll talk a bit about the high value capabilities and use cases what we're seeing for digital transformation talk about how those tie into the digital operating model and what are the key elements of that and what are the use cases that we're observing especially in the in the context of covet and some of the need for visibility and adaptability right and to the end i'll talk a bit about how do we measure value especially for planning solutions uh what are our value drivers how do we measure value uh value potential for these customers and maybe some customer stories as well depending on how much time we have uh i'll try to cover as many as we can okay so that's a quick summary of what we plan to cover today um and uh maybe let me start it off just so that it doesn't get too power point-ish i want to try to intersperse certain uh concepts through our user interface as well and our application and solutions okay so maybe let me introduce that concept what are we trying to get to and and where the gaps are so i'll take a simple example of google right so today in in in your experience in consumer experience if i need to search for anything i can go to google and then i search for for example let's say i want to search for seminar and google is able to find that answer for me right it comes right here it gives me a list of suggestions topics and then it takes me to that specific page whereas the same concept when you map it to enterprises is extremely hard right there is no google search within an enterprise context right when somebody wants to understand what is the status of my order they have to go through four or five different systems to actually understand where that order is what is the status of that um who's the supplier for that order it's not as simple as what you what we've become accustomed to right in the in the consumer world so i think what we are trying to get to is that so i'm gonna switch quickly to um one of our platform screens that we have and look at what would that mean if we were to translate that into an enterprise context right so for example i want to understand uh status of an order so i can type in an order let's say i have an order 192 or let's see order 192 time i have a list of orders but what we want the system to be able to quickly find out and this is where i'll introduce the notion of an enterprise knowledge graph where behind the scenes it has mapped out your entire knowledge graph of your enterprise and we're trying to quickly search for information right so for example as i start typing information such as what are my top five due by let's say gross revenue so you'll see as i'm starting to type what you what you're observing is matches to elements within the enterprise knowledge graph right so it knows that skew is an element uh gross revenue outlook is a measure or a kpi that i have access to so it's trying to build that knowledge graph and it's able to pull that information in real time right what we're trying to show is can i quickly and easily access the knowledge graph to get the information out similar to google and try to present this information right so what we were trying to start off with order 119 192 for example this is one of the orders that i have concerns about or i have received an email about that i want to quickly check so i can go and look at the order properties what's the order from but i can also navigate to the screen to understand the status of that order right the supportability of that order and this is really where the power of the knowledge graph and enterprise uh digital transformation comes into play is the notion that i can go to any order understand what is whether it's fully supported or not where are the constraints right so right here i'm seeing that it's unable to fulfill 443 units of that order and then i can look at what are the root causes root causes is because of lead time before current and so on so this gives you a digital representation really of the knowledge graph and then it's able to manipulate the graph and help you understand where the issues are and so that's really what we're talking about when we talk about digital transformation is taking some of these concepts around being able to do google-like queries on the enterprise data being able to look at uh physical representation of your supply chain so another example would be let's say you have a physical representation of your supply chain this is just a simple uh screen that i took from one of the images out there but if you have an assembly line or a factory line and how do you represent some of these elements into a digital twin of the enterprise right and that would be the other concept that maybe we'll touch upon towards the end is how do i represent a supply chain network but being able to address different levels of detail right so i can zoom into my supply chain digital twin to understand what does my supply chain look like but i also want to understand how do i look at details within each of those nodes right so i can zoom into a plant location to understand what resources are consuming what product lines what are the assembly operations in terms of what are my raw materials what are the resources being consumed but this also allows me to then quickly navigate from here to look at more details about if i want to maintain my bill of distributions look at my priorities of the lanes etc all of that is being managed through here so this is really what we're calling the digital twin um and trying to map out really the physical supply chain to a digital representation so that it's easy for us to easily quickly understand where the issues are provide visibility to the extended supply chain right going all the way from your suppliers uh to your supplier suppliers and so on um so those would be concepts that i'll touch upon as we go through the presentation but what i wanted to give you a picture is where we're heading to as part of this session we'll try to uh touch upon all of these topics is what i was at the end i'm hoping that you'll take away how we what we think about digital transformation and why it's important and some of the key topics are on digital operating models enterprise knowledge graph and and the few things that we talked about okay so hopefully that gives you a picture of where we're heading to uh to put it in context and then let me start going through some of these slides so that we can let me start my pointer here so then i'll start to go through these topics and then we can see how we if you have any questions on those so who's online i think in terms of what we're trying to uh be is we are new kids on the block in a way we've been around shorter than some of the other competitors but our value proposition is to create the digital brain of an enterprise right and what we're talking about digital brain is uh really this interconnection of integrated business planning platform uh for transforming these specific solution areas so as you can see we are focused on planning planning is our in our dna it will be around your commercial planning or revenue planning financial planning uh ibp snop plans and supply chain planning right so those would be the core focus areas for us in terms of the platform uh what we're calling integrated business planning platform and really at the end of it is how is it getting powered is through this enterprise knowledge graph and digital brain some of those are these elements that i'll touch upon as we go through the various slides here now in terms of customers um we do have a breadth of customers like i mentioned it's a platform and what that means is really it is uh it is uh configurable it is not set in stone so there are elements of the platform that you get as part of pre-packaged solutions but it is configurable and that's the reason why you see a breadth of industries being represented all the way from industrial manufacturing like caterpillar to brand manufacturing nestles and and a1 etc we have retail accounts some of the largest retailers uh walmart's and and starbucks etc on the solution we do have telecom customers we have oil and gas customers and pharma and so and life sciences etc right so it's a it's a breadth and then that's where we're trying to position the power of the platform being able to adapt to these various customer business problems and that's really the kind of business that we are in we love the challenge of looking at various industry problems and trying to configure their solutions so let's take a small business case here maybe just to give you an example of what kind of solutions we're looking at i'll give you a quick example of starbucks in terms of what what they were uh what challenges they were running into and what kind of solutions we were looking at so that you get a feel for how we're mapping the business process to our business problems into our platform itself so the the key problem that uh starbucks had come to us with was really their store managers and baristas they were spending a lot of time uh trying to create replenishments and manage inventory at the store right most of the time they were in the back room they were spending time trying to create those orders what's my next order going to be and taking care of the fires whereas what they really intended or their objective was that the barista should be actually building customer relationships they should be in the front of the store they should be interacting with customers coming to know their customers and what they need um so it's more of customer experience and they were inundated with doing the back ordering and and some of the backend tasks so that was really their main ask was around how do we free up their time by trying to help them uh with this automated store replenishment right so the initiative was more around asr where they wanted to but there was some complexity involved it was not a simple store replenishment because they do have to consider kelp and store constraints i have limited shelf space at the store i have a lot of wastage if you know from looking at the bill of material and recipes that they have where they have a lot of leftover milk and and cartons and and all the ingredients that go the syrups all of them there was a lot of wastage there's a lot of labor costs at dc's for example when i'm looking at kitchens the sandwiches are really assembled in in what they call kitchens right and those are not in the store and there is labor involved to assemble those sandwiches as well so they were looking at so they were looking at bunch of these complexities in terms of solutions for them to help address both the replenishment problem but also to consider what are my local store events that are happening right so the other problem that they had was in addition to this the replenishment they also had the element of incorporating drivers to improve their store forecast right and this was a use case where they were looking to add market and local indicators by that what we mean is uh what is my temperature because temperature has an impact on my local store traffic uh weather and temperature the severe weather events then obviously the store traffic goes down or if the temperature is cold then my category of warm beverages sale goes up compared to cold right so the correlation of impact of temperature weather and the impact on the category forecasts as well as their store forecast being able to consider local events for example there are uh pta meetings that happen near the store there is obviously going to be an increase in my need or if there's a local football game for a high school football game there is a lot more traffic at my store right so those are things that that is hard for systems to understand today because most of the systems are looking at historical data whereas some of these are leading indicators i know that next week there is a local pta event because of which i might need to order some more sandwiches so that's really the context around this what are the challenges around creating this automation for forecasting being able to consider a lot of these leading indicators and drivers and taking them to the next level in terms of automation and planning so maybe just to show you what this translates to i'll give you an example on our platform itself uh so let me just go back to our screen here uh so one way to for us to navigate to that uh session let me continue from everywhere before with this order so for example i have a set of stores and i just want to look at my store total forecast so i can start typing that in here in terms of trying to search so total forecast is a metric but i also want to uh for let's say uh store one and let's say q2 and for a particular time period right i want to look for my quarter in this data set the quarter was 2019 and then by day so what this is trying to do is it's going to look and try to match what are all the elements in the knowledge graph i know store one is an entity quarter is an entity days and entity and total forecasts right so based on that it's going to try to automatically render this information for me in a chart because it understands almost nlp like it understands uh by day meaning that i need to render it on a chart with day as the x-axis and so on right now one of the things you can do here is to further extend this right and i want to say in addition i'd like to understand temperature forecast and so i can extend this forecast right so here what we're trying to do is in addition to store forecasts i'd like to overlay my temperature profile on the forecast as well right so then now i'm adding another measure so one thing you'll realize here is it is not something that's predefined in the system a lot of the existing bi tools you'll have to pre-create these reports and store them and then when you need to query something ad hoc then it's out of date right you don't have a report to support it but as here what we're trying to get to is change the paradigm a bit trying to search in real time understand what the knowledge graph has behind the scenes trying to query that and pull out the information in real time but also from here navigate to those specific entities right so i can navigate to my uh ml forecast model for that particular store i was looking at store one and then this gives you understanding of what the model itself is so you'll notice that in addition to just my historical drivers um in this case we also have uh we also have other causal factors right right temperature weather you have your marketing initiatives uh product reviews etc those are drivers that impact uh what my products are going to sell right and so we're trying to pull those in and being able to run these ml so these have become very easy for us to do now being able to put in a lot of variables and we're trying to help the store managers and the starbucks team understand what are the driver correlations right so you can understand maybe seasonality is the most impactful driver for this particular store for this particular product line right so be able to create that correlation for them so they understand which drivers to pay importance to now one of the the things that they wanted us to do is being able to translate this information so i'm looking at a store forecast here and then one of the things they had challenges with was store ordering right because there are usually special store orders that are not captured in the system and they wanted to make it easy and one of the easy ways for a platform to accomplish this is to provide various interfaces for the user right so think of this as the central planning team that is creating these replenishments for the store but i'd like to get inputs from the store manager right as well so let me switch that to a different uh ui here i'm just going to switch to my mobile interface so here you'll see that i'm actually looking at uh my phone and i have an app the online app on the phone so think of this as a store manager workflow where the store manager at the same store right i'm looking at store uh store one and then this is my current working view and so on but i'm looking at let me just switch my filter my dates to five eight so i can set up a replenishment or a special order on that day right so as a store manager i can quickly look at the information here and if i wanted to create for example a special store order on this date right i can say i need 50 units additional ordering for this particular store and then why am i ordering that that's really where the collaboration comes into place so the built-in collaboration features where i can come in here and type in my comments to say there is a special store event on this particular day for to support a local pta event right and this could be orders for my spinach cage free egg wraps or my cappuccinos etc so i need to create some of those additional orders needed as ingredients for those products right so based on this input that i've provided as a store manager then immediately the the central planner gets notified right all he needs to do is once he gets notified he can refresh his screen and you'll see immediately see the impact of the store order that was placed by the uh store manager right so i'm seeing this 50 units that i entered on my on the store my store manager's mobile phone show up here and then i can understand why so i can look at the store pulse to understand what is going on so there's a special order to support my event and i can collaborate on this so think of it as almost facebook and twitter like collaboration where i can respond to these comments um is this confirm and this becomes an interactive and and uh and it gets archived in the system right so there's no more uh papers and emails floating around you can do the collaboration on the system you understand why you could capturing the reasons behind some of these demand uh surges and so on and that really is really the power of the platform that we're trying to portray here is how do we take some of these genuine use cases that customers have challenges around store ordering challenges around creating these automated replenishments and so on converting them into uh simple business problems and trying to solve them uh that's really where we uh we pride ourselves in in looking at some of these genuine use cases and solving them okay so that was just a quick example of starbucks just one such use case we have a number of complex use cases uh walmart's and google has a lot of complexity in their supply chains uh abner has a lot of stuff going on as well so there's a number of these use cases if you need more information we can uh probably share that with you but each of them has complexity and different types of complexities as well okay so with that let me just switch over to uh finish up this topic on who we are this is a sort of a bird's eye view of o nine solutions again we are headquartered in dallas texas that's where we we started the company but our heritage is from i2 solutions so if you remember i2 technologies uh if you remember uh sanjeev sidhu and so on we had started a company back in the 80s which was a pioneer of supply chain uh planning right and so we've continued the same heritage into online solutions some of our founders sanjeev siddhu and chakri both of whom have spent a lot of time in the industry and and spent a lot of time with customers so this really here this was a reincarnation of i2 technologies but not just focused on supply planning here we're trying to focus on uh integrated business planning which includes all of the solutions that i talked about and then really what we wanted to differentiate ourselves on was uh cloud-based solutions so it's a native cloud-based solution it was born in the cloud most traditional companies have tried to adapt they were all on-premise solutions that are trying to convert to cloud whereas we started from scratch so we had the advantage of building something from the cloud and trying to bring in a lot of learnings from talking to various customers over the last 20 30 years and trying to bring in all of that into uh an in-memory uh data big data platform where we can address some of the challenges of planning uh analytics and and uh and operations as well okay so that's a quick summary i'll just talk a bit about this in the next slide where we are today is uh talking about the leaders quadrant which is uh which is really uh something that we are proud of you think look at where online is today it is positioned as a the most visionary in the leaders quadrant uh even though we've been on the market for a short time right the platform is about four or five years old and that's when we made it public and and started getting customers on the platform but based on what gartner is seeing we have the most visionary ideas and we are the most disruptive and innovative in the space trying to bring in a lot of the new ideas and these are things that again a lot of things that you guys work on in academia and things like that where we'd like to take a lot of those learnings and ideas translate them into adapt them to business challenges as well okay so with that let me switch over to the next section so that was an overview of who's online let me switch over to the next section here so in this section what i wanted to cover was what is digital transformation uh there's a lot of buzzwords out there and a lot of executives who get confused so with that let me just open this slide here it's really around uh these keywords that you keep hearing right there's artificial intelligence digital transformation machine learning big data analytics there's digital assistance uh and so on and and executives ask us that question what does this really mean for us right and now that kobit has really what we've observed over the last few months is that these were things on the executive's mind in terms of digital transformation but it has really been accelerated now from based on last year's what happened with the pandemic that has caused an acceleration for executives to start moving the needle right now they're forced to you'll see a lot more new roles in organizations now you have the chief digital officer or chief digital transformation officer those are roles that we're starting to see more and more from the other traditional executives like chief supply chain officer or chief revenue officer now we're starting to see some of these new roles and they they are really interested to understand what this means how does it work for their company and and what can they do about it right so what we're trying to help them understand is this in terms of how do they incorporate some of these uh through the inter integrated business planning process how how do they incorporate some of these into the planning process itself okay so let me just talk a bit about some of the pandemic learnings what we've observed over the last few months again these are things that we have seen it may be different for different industries but in general what we've seen is uh is some of the uh concepts let me talk a bit about each of them i think we have about five of these what we've been observing one is around demand variability right and this is something every one of us has observed right the product makes shifts especially around essential products so when kovitt hit a lot of people were hoarding they're buying up on essential products and and the sales of all the other categories dropped significantly right nobody would have predicted that if they don't have the right systems uh most businesses have trouble uh giving visibility into which essential products are going to be in demand which are not right so there was no opportunity for them to plan ahead and this is where we're trying to understand what are some of the capabilities needed for businesses to stay ahead right so this was one the other one was around channel shifts right a lot of and this has been happening for some time uh people buying online but last year there was a dramatic shift like nobody would have predicted that the shift would be that dramatic where everyone everything is moving online right then we have this regional variability being able to understand which regions are in lockdown which regions are opening up which countries are going to have locked down which countries are opening up there's a lot of complexity here especially with the fluid situation of kobit right as it was surging there are new strains of the virus that's impacting that requires additional closures things like that being able to incorporate some of them and obviously those temporary spikes we talked about so that's really on the demand side where we observe some of these challenges that that most of the companies were facing then on the supply side similarly there is supply challenges because suppliers factories were shut down because of covid or the capacity was reduced significantly they are unable to understand uh which of my manufacturers suppliers can can fill in for my uh needs across the product lines so there were challenges around all the way from your your own resources and your own factory lines slowing down or stopping or coming back online to visibility across your entire extended supply chain because you don't know if my tier 2 or tier 3 suppliers have issues then how does that translate into issue into uh constraints on my supply chain right so those were a couple then the third one was where we were seeing a lot of asks around these uh uh war rooms right so there was a lot of war rooms being set up just to address these challenges every day right every day it was a fluid situation but being able to understand the channel shifts what decisions what actions to take short term what actions to take long term how do i bridge the silos between the demand supply uh being able to do what ifs being able to run real time uh planning cycles rather than these weekly batch cycles right so some of those were being challenges that we observed and getting accelerated through the pandemic itself the other couple is more around collaboration so we did see like we are doing right now this is a virtual conference or a virtual seminar we've seen a lot of virtual conferences going on similarly in in a lot of the companies we saw that uh they had to go digital so instead of in-person meetings for snops or resource coordinations or your snoe etc all of those meetings went digital and they were doing collaboration online through zoom meetings and web x's etc so that's sort of becoming the new norm right so they were forced to move from power points into the online tools where i can see and share information in real time right what i am seeing versus using there should be no lag between those we like to share the same information so those were some of the challenges on the uh collaboration that we've observed in terms of learnings and then the last one is around digital operating models now this has really been uh a trend that we've been seeing where a lot of the companies are trying to uh make sense of all the data that's online right so now you're starting to get a lot of data but they're not set up to understand the data and use that data so which is where we what we're starting to call these digital operating models where companies like amazon's and ubers and airbnb they have a leg up in terms of how they are set up to cater to these transformation changes and the pandemic and huge shifts in in demand and supply right so those were the 405 we'll use these as sort of the basis to help you understand what capabilities are needed in the platform to to address some of these challenges okay so maybe just to put this in context maybe i'll focus a bit on the digital operating models and what we mean by that uh i'll just switch over to the next slide really what we're seeing is this right most companies are traditional operating models what we're calling them so think of them as how they've grown from their dna is being a retailer or being a brand manufacturer or being an industrial supplier that has been the core and then they have over time try to add some software capabilities right to address planning and so on but those have been piecemeal whereas what we're contrasting that with is some of the new age companies like ubers and amazons and googles etc or airbnbs stitch fixes all of those companies where they have a software dna right and they have created software platforms that is then applicable to their business that they are in now there is no given that they have the software dna you could always say amazon can can apply a uber problem and solve that or uber can become an amazon right so they're in a way they're interchangeable because software is driving everything right the tesla they thought of it as a the car is really a software and then they built the car around the software right so that gives them a lot of flexibility in terms of ease of changing their models adding new features and capabilities in real time so they can continuously provide updates to their models and new features so if i want to add a self-driving feature it's just a matter of turning on a software update right rather than a whole physical redesign and i have to buy a new car so traditional companies would have had to do that whereas these guys are thinking different right and that's the mindset we're trying to bring is that digital operating models are the new age and you have to transform from traditional to digital if you want to play so a lot of these companies like walmart's and nikes etc they have started to think along those lines they have uh big it arms that is building a lot of capabilities in software but there are a lot of companies that are not there yet and which is where we are saying you need platforms that can help you transition from traditional operating models to digital and we are here to help you accelerate that transition right now one of the use cases where to understand what this really means we took an example of amazon right what is their advantage of the digital operating model so that most of us shop on amazon so we're familiar with what they do but really it's around these two key pillars right is around the digital customer engagement and the digital supplier engagement right so customer engagement is really where they're collecting this customer knowledge right they understand what is the purchase intent of a customer what the demand levers cause sensitivity to the purchase what are the product feedback ratings they're collecting a lot of information about the customer but they're using it right so they're taking the information they're using their own ml analytics their demand forecasting consider some of these drivers to then allow us to shape that plan right so i can shape where the product should be placed on the page i can shape my pricing plans i can shape which one should i be promoting cross-selling up selling right so all of this is happening digitally almost in an automated fashion right once you get information and building the customer knowledge the algorithms take over they're able to process that and recommend how do i shape that how do i improve the chances of sale and some of those capabilities right so it becomes almost a digital um model here similarly on the supplier side they're collecting information about suppliers what are they good at how reliable are they how's the quality what capacity do they have who are the alternate suppliers so a lot of the knowledge base for suppliers and they're leveraging that to make uh collaboration more effective right how do i make sure i can commit to an order on time how reliable are the forecasts uh do i have visibility into inventory because it gives you all the information on the app right you can understand uh where how what product how many units are left to order when will i get my next order in etc all of that information really think of it as everything being digital right whereas it's very hard for traditional companies today to even understand what their customer intent is so that in a in a nutshell is really what we're trying to uh map out is how do companies move to something like this where they can understand uh both the customer and supplier sides of the equation right so our in our in our opinion what we are defining digital transformation of specifically around the integrated planning and operations is really these key four or five things right connecting customers your channel partners suppliers planning processes and business operations right so those are the four or five things that we want to tie together through this digital operating model where everything comes together and then we use ai and digital technology to actually provide these capabilities to enterprises right and these four or five capabilities where we narrowed it down to is one is around seeing your risks and opportunities earlier and this is where the market knowledge and leading indicators uh the story that we were talking about for starbucks comes into place and then enhancing that with intelligence right you'd like to make sure that these are driven by knowledge you apply ml aiml to the mix to to give them better decisions then the third one around processes enabling better collaboration right so real-time collaboration uh rather than having to set up these meetings and call 20 different people to understand what's going on collaboration live on the tool to enable your decision making and execution processes and then finally on automation right making sure a lot of these mundane tasks are automated and then you have the digital assistants to support you through that process okay so that was really what uh what we mean by digital transformation but there are challenges and what i wanted to touch upon is the current state as well right so today uh businesses are running it's not like they're going to stop and do this transformation but there are challenges and those are the challenges that we want to take to uh to the executives to help them understand why because you have your snop meetings today there's a gap between forecast and plan and executives are asking the same question right why is there a gap uh why were they demand surprises why were there supply issues why didn't we know earlier why didn't we act faster so those are kind of questions that are continuing to be asked and it's primarily because there are no systems in place to provide that answer right they have to go to 10 different systems to to figure out what happened do the root cause analysis and come back with an answer by then things have changed already right and really if you boil it down to what's causing some of these is really the overlay of the land right if you look at any traditional company their traditional operating models there are a number of systems right again these are necessitated because you do need your factory scheduling is a tool you need transportation management you need commercial planning tools financial planning supply chain planning so the number of these tools which are siloed right because they don't talk to each other they're on separate sets of data sets they're on separate cadences each of them has their own knowledge base and eventually the planners resort to using excel and powerpoint so a lot of them is not even in the system they are in excel right so there are a lot of challenges around information sharing any changes that i have an impact in my factory operations by the time it gets to my p l impact it's already pretty late right there's a lot of latency in the system and that's primarily because decision making has moved eventually ends up being in excel and powerpoint that's really good the go to go to tool for most planners and having such a system in place and this is really the reality of enterprises will be seeing and what we observed is around the value leakage right it's almost like 10 to 20 million dollars in value that is being lost because of some of these issues that we're seeing from slow silo operations and and the spreadsheets that are all over the place right and if you were to look at the story of typical implementations what we've based on our history with what we've looked at it boils down to this low adoption back to spreadsheets right and even though you spend millions of dollars implementing uh these erp systems or these solutions what we've seen is first of all it takes a long time right you take six to eight months to create a brd itself what is the design document and requirements and then you start implementing and then the user sees new paradigms that they're unable to adapt then by the time your business has changed right let's say like covet hit and you you need to be able to make some changes in your business models or for example you have to add you're switching from outsourcing to insourcing or you're switching to contract manufacturing uh things like that where the models need to be flexible right and some of these models are inflexible so what they implemented by the time they look at re-implementing that with huge cost and time implications uh the planners are back to their safe zone right they go back into excel so really this is the story of a lot of implementations and enterprise solutions that we've been observing where we want to take it to is trying to bring a consumer uh experience to the enterprise right so i talked a bit about the google like so i tried to map out a few things here the google-like experience where we looked at fingertip visibility that you can easily get through consumer systems whereas extremely tough in in in planning system in enterprises right similarly collaboration right real time today everyone's on twitter and facebook and and um whatsapp those are real time collaboration tools whereas in enterprises they're stuck in silos right you call for meetings uh there's a lot of alignment issues that happen during meetings and so on and a lot of time is lost there similarly if you're looking at visibility and prescription so if i create if i want to go from place a to place b it tells me the best route to take in real time right if there's traffic on the route i can use google maps to understand it will give me alternate options and so on similarly there's nothing like that for enterprise tools right where i can look at plan misses or forget early warnings into where some of the bottlenecks may be and then we've obviously lived through amazon's and netflix the learning systems netflix recommendations that comes our amazon recommendations based on the browsing pattern the browsing history that you've had whereas in enterprises it's typically tribal knowledge where a lot of the knowledge is heavily human biased in the decision making and so on similarly your demand supply match right in real time uh for example when you order something order an uber you know exactly uber does the whole demand supply which drivers are nearby uh what's the demand in that area how do i quickly find the right driver or the right for you and being able to understand where the driver is all of that in real time think of it as that experience is missing an enterprise today right so there's a lot of firefighting that goes on i don't know why the demand supply is off i can't find that in real time i need to wait for a day or a week to understand if i were to rerun the plan what does that look like right and then finally around digital assistance where a lot of these voice based assistants that we have today i can ask alexa any question and i get an answer whereas in in enterprises today because they're all spreadsheets powerpoint driven it's extremely hard right so this is trying to paint that picture really of what is happening today and where the the opportunities are right so you can see there's a lot of opportunities here to make enterprises uh benefit so even small moving the needle a little bit really helps them significantly taking the experience from what we've learned from consumers bringing that to the enterprise is really key right so with that let me switch over to uh i think we are at 10 20. okay so i'll spend maybe a few more minutes on what these high value capabilities and use cases are so that gives you a background right what i've covered so far is really around uh what we mean by digital transformation and what is the current state reality right where we want to take this now what i wanted to touch upon is uh what are some of the key use cases and how do we think of bringing these together right so i'll talk a bit about the the digital transformation the key pillars that we're seeing um in the market more and more what we observing is there are these four key pillars four or five key pillars in terms of what the chief digital transformation officers are looking into right one is understanding the consumer the customer engagement is becoming extremely critical most of them with the direct to consumer now a lot of the companies have moved to d2c model where they're looking to understand consumer behavior what are the customers doing what are the purchasing behaviors and not just in retail context but also b2b right how am i what are my opportunities with my b2b customers and so on right so a lot of the data again what this really is translating into is now you have more and more data contextual data about your consumers about your customers that you need to understand and process right and uh then the other pillar is around product innovation right so you've seen this transition happen every device that you have today has some sort of sensor in it right so whether it's your your toothbrush or your roomba or your smart watches etc they have iot embedded they have embedded software that is capturing a lot of information right and this is really where again it's boiling down to data right so you're getting a lot of data from devices and your products and you need to understand what to do how to use that in planning right and decision making how does that help you make better decision how does understanding my customer help me make better decisions similarly you have your automation in your supply chain right your factory digital factories your automated warehouses where you're getting a lot of automated information through iot devices and so on uh your sensors you know when your factory is going your products your resource lines are going to go through maintenance uh you know when your shipments are getting delayed etc all of that information is coming through now what do we do with that information right one of the quotes that i read is in the near future the most precious commodity is going to be data right and that is really the crux of what we're trying to do you're getting a lot of data now how do you make sense of the data and how do you plan and make better decisions with that data and that's really what we're positioning as the digital brain so let's move to the next slide what we're positioning as a digital brain which is really the core engine of your enterprise right so think of the brain that is uh the human brain which is taking a lot of information and processing that knowledge right and the analogy i like to use is converting that data into knowledge is really the key right what a human brain does for example i may have visited 20 different restaurants over the last couple of weeks i may not know exactly which restaurants i went to but i would really know which ones tasted the best or the worst right so what it is really trying to do uh without you realizing is it's taken all of that data all of your site visit data converting them into knowledge so now you remember it's converting that into memory right that i which are the best restaurants that that i went to or the worst restaurants similarly what we're trying to position here is the notion that you have you're learning from you're getting a lot of market data in terms of external drivers your customer information a lot of that coming into what we call market knowledge stream converting that data into knowledge similarly your demand knowledge you have your demand drivers that are coming in your point of sale all the information that is coming in what is my forecast accuracy and so on all of them being converted into knowledge right and same thing with supply side so our positioning here is really around taking a lot of the data converting them into knowledge because that is what is eventually required right for me to be able to make better decisions i need to know which supplies are reliable right rather than going and figuring out which supplier should i talk to system should be able to recommend the best suppliers right for this problem or which uh which demand drivers do i need to look at which region should i be looking at for upsells or which customer should i be upselling or shaping my demand to all of those decisions come through the digital brain and how we are looking at it is this right these four pillars coming together and the digital brain is really the core of helping you plan and based on this we've identified maybe what are the key four or five things where we we plan to focus on again based on what we saw in terms of the challenges right what we're observing the five key challenges and the areas that we feel the capabilities need to be one is around real-time market knowledge right and this is extremely critical to understand what are the leading indicators of demand uh how do i look at my external data internal data and connect with dots in my supply chain right all the way from consumers to my supply side the second piece is then taking that knowledge and how do i improve my forecasting now a lot of manual forecasting that happens or your traditional time series based forecasts which are based on historical data that is really backwards looking right what we are trying to position here is uh think uh leading indicators so i know what's happening in the future based on these external demand drivers i know that there is uh let's say my gdp rates or i have a new election that just got over but if i were looking two months ago i know there's an election coming in what is the impact of that on my forecast right things like that where you could have correlations with various events that are happening in the supply chain uh those are things you want to incorporate into your forecast the third one is around synchronizing your planning right how do i synchronize my supply plans and demand plans being able to run these what if simulations on the fly right and that's the critical piece i'd like to not just understand my change in forecast but can my supply chain support that forecast right and that's why we want to be able to run these simulations what-ifs in real time but also understand what is the impact on the cost right how much even if i'm going to support the additional demand how much is it going to cost me what additional cost operations do i need to consider is it going to be profitable so being able to consider the the financial implications of your plans is extremely critical then the fourth one is around collaboration right we want to bring a real-time collaboration on the tool between suppliers and customers and the planning system right it's a real-time communication where you really shouldn't have to do this offline you send excels to your customers or suppliers and have offline meetings they should be all on the same system itself right and finally around continuous learning right so how does the system learn continuously all of the actions that you're taking you're planning uh how do i improve the processes and so on right so that's really the crux of what we're trying to get to here in terms of uh continuous learning so based on this let me spend a minute or so how this comes together i know this is a busy slide so i won't spend too much time but what i wanted to highlight was there are five or six different elements that come together to make up this platform right we talked a bit about data so there's a lot of data sources that pull in data it could be from your enterprise data sources such as data warehouses and erps you could have your market knowledge streams supply chain knowledge stream these are external data sources that come in but all of them are feeding this enterprise knowledge graph and this is really the brains behind everything it's trying to create the absorb the data and this could be again big data sources uh where we have to store all of the data but then convert them into these knowledge models and those knowledge models really power the the analytics right so you can do a lot of analytics only knowledge models and then use them for all of your planning processes all the way from ibp to supply chain to revenue so there could be a breadth of planning processes and then take that information and collaboration right so engage with customers uh you have different types of customers or engage with suppliers your tier one tier two collaboration etc and then also integrate with business operations right these are where you talk about these fringe operational folks so the planning is almost central but you also want to get inputs from finance your logistic operations or your data scientists your account sales who have certain actions they're taking uh when they're talking to customers for example how do i shape the demand they expect certain inputs from the platform right in terms of helping them guide their decisions but also get the inputs back from these operations into the platform right so this sort of thing in a nutshell again we could spend a lot more time here but uh what i wanted to quickly talk about is a lot of elements that come together to create a platform like this in terms of art of the possible because all of these are critical right critical elements that that need to happen for us to enable the digital transformation in enterprises so maybe let me spend a couple minutes on the enterprise knowledge graph itself so that you guys get an idea of what we mean by the ekg really this is the ekg model behind the scenes in a way where we're trying to say there is a number of relationship types right you could have hierarchical relationship between various entities and also your network graph relationship and what this brings together is really the two which is trying to correlate the hierarchies and networks in the same data model right and that helps us really create these knowledge models around your forecast knowledge models or assortment knowledge models or competitor internal intelligence customer intelligence models etc the number of these models get created but what we want to also highlight is these are not static right think of it as the brain that is always evolving as more information comes in you'd like the brain to actually uh take over and extend the models and make it so that it's flexible i shouldn't have to bring down the system to actually add a new model to the to the mix it should be automatically uh able to extend the graph and that's where the power of the graph comes into place for us because we're able to easily add nodes into the graph without having to bring down the system and these should happen automatically what we're envisioning is if an email comes in for example i have a new competitor x that has a product launch uh next week right so now let's say competitor x was never in my system i didn't know about competitor x i shouldn't have to go into my erp system and add competitor x and then uh figure out an etl transfer of that data into my system let's say the email comes in it automatically figures out competitor x through the email tagging and that's what i was showing you in the nlp search there's a lot of tags that get created competitor x becomes a new tag and it's related to product y for example so then that association happens automatically through the knowledge graph right competitor x knowledge and product y those get related and i know who's the owner for product y right i have a relationship to the owner as well which is the people so i know any email that comes in from external systems automatically gets processed it gets associated with the product and it sends the email to that person that's in charge right so that's really what we're talking about evolving the brain rather than a static representation new models coming in get part of the brain and the brain evolves right and you're starting to make it richer and richer as as it goes now there are some features that i really don't want to get into at this point it's a bit more detailed around the need to have this big data store behind the scenes that's integrated and the extensibility real-time apis are critical because now you're getting real-time information and how do you easily access that data and we talked a bit about converting data to knowledge and we're just going to skip that with that let me just skip a couple of slides i know we may run out of time so i'm gonna try to hit some of the highlights uh we can skip the challenges and this one what i wanted to touch upon was this one which is uh some of the capability gaps and use cases that we're seeing more and more so and i just wanted to highlight the top five in each of these categories uh these would be in demand planning snop ibp and supply planning right so in demand planning the ones that we typically run into and where we see a lot of issues is leading indicators which we talked about right really you see a lot of surprises today in in business plans and demand plans because they did not consider the fact that there was a competitive product launch let's say two weeks down the road or there is a severe weather event that's gonna happen in new york next week right uh things like that where it's drivers that might impact you demand not being considered that's a big challenge that we see today similarly visibility into your initiatives that you have so i have my own product launches or i have my own marketing let's say super bowl advertising event so this year a lot of companies have dropped that from from there so pepsi is not going to advertise or focus not going to advertise and those are things that need to be considered because my forecast for that same period is going to be different right what i used to get the uplift from marketing some of these marketing events is not going to be part of the mix so then because they have different systems it's a big challenge for them today similarly in b2b we see a lot of challenges incorporating your opportunity pipeline into your forecast right and so on so those would be the four or five top things that we see around stat forecasting a lot of them not able to use aiml techniques and so on right in snop i say maybe the key one is around conducting meetings in powerpoints and this is really a big big productivity drain right by the time companies spend the time to create these powerpoints for an snop meeting and they do it monthly so it's a monthly meeting but they spend a week actually collating the powerpoint itself and there's a lot of people there trying to create that and by the time they get to their monthly meeting the data has changed right but they can't go back and edit the powerpoints because they've taken snapshots and images and so on these are 100 slide decks that they create and it's very hard for them to update them right so that's one of the biggest challenges we see in snop to make it uh it's very hard for them to react to to situations in real time and then most of the repo it's mostly reporting based and they're not able to do gap closure actions right so if i say i have a new opportunity for a new order at customer x there's no way for me to say that i can i support that order or not right or if i need to close the gap to my plan what are the initiatives to close that gap being able to do some of those what-ifs is really where the challenges are similarly on the supply side a lot of visibility i think what i've shown you in the ekg is really uh mapping an extended enterprise right so it's not just within my enterprise but being able to look at tier 1 suppliers or tier 2 suppliers understanding where their factory locations are and where the gaps are going to be in terms of supply issues or supply constraints uh identifying some of those so those would be some of the four or five top use cases that we see typically in enterprises right so let me touch upon one or two of these uh one of maybe one of each to give you a flavor for what we mean by uh by some of these use cases and how it translates into our tool as well so one would be demand planning opportunity which is really reacting uh where we're seeing maybe in terms of maturity journey most of the customers are in this space right where they're looking at orders which could be orders or shipments they're looking at that again it's all historical data they look at order shipments and try to project what my forecast is going to be what we've seen the next level is some customers are sophisticated enough where they're moving from these sell in what we call selling as the shipments and orders to sell out based models where they're looking at maybe customer sellouts right so the point of sale at the store there are a lot of stores that that provide point of sale data channel inventory and so on and moving that to needle to the driver based models where you're able to collect a lot of these external drivers right and these are again areas that there's a maturity involved because a lot of folks don't know how to collect the data and what to do with the data and that's where we're trying to uh provide some capabilities of doing both of these right so let me just show you an example let me switch here so i'm going to switch to one of the knowledge hubs that we have where we are collecting a lot of the information again a lot of this is public information but we are able to collect the information from for example google mobility so it gives you these trends information to help you understand i know that during uh march april of last year uh people were not going to grocery stores so you'll see the trend really reflecting that most people wear in residences which is why this is higher compared to all the other locations right so a lot of this data is available publicly and and it's available for us so we've been collecting this information through various websites your demographic trends your uh which cities have what kind of trends in terms of number of elders number of dependents and so on or events right so the events in terms of holiday events that might impact my sales being able to collect this information sporting events etc but uh so what i'm trying to show here is just a collection uh data collection for the market knowledge right so be able to pull in this information but how does this get used maybe let me show you a quick example here what we're talking about the aiml and the black boxing the outputs right so this is an example of a forecast that we see historical forecast it's going to be extremely hard for us to predict what the forecast is going to be for this but if you were to break it down further right so we can start looking at what are the correlations does temperature have an impact right so you can overlay temperature trends for the same location for example in walmart for that region you'll see the correlation is pretty low i see the peaks are different from what i'm seeing here so then you can check that out and say what if it was number of days on promotion yes which has an extremely high correlation every time there's a promotion there is a peak so i can understand that but then it still doesn't explain why this peak is so much higher than this peak right so then i can start looking at specific types of promotion events in terms of vehicles whether it's a shipper or an aisle promotion or a [Music] shelf on the shelf etc so you can start overlaying this this is sort of instead of having it as a black box the planners actually have the ability to overlay so they understand exactly what event caused some of these spikes but still it doesn't explain really why the the big spike happened right so what we can then do is look at some of the other drivers one other example would be is it a weekend or is it a holiday so holidays does seem to have an impact you'll notice that new year's and christmas has some correlation to the bigger spikes but it still doesn't explain the actual spike itself and and that one is actually related to pricing right so if you look at the the price uh driver you'll notice that not the price and the discount so if you look at discount it actually correlates pretty well so there's one one time when they had a 74 discount on the product versus their standard discount is 49 cents so either it was a one-time promotion which caused it but just to give you an idea really what we're trying to say is taking in a lot of the information of the market drivers leading indicators correlating that with your forecast and then looking at the ml drivers right it tells you that chrome discount is really the highest importance in this uh segment which is what you need to focus on right so anytime there's a small discount change there's a big impact on your on your sales so that would be an example of how we think about bringing in some of these external drivers into the mix and then trying to influence what the forecast needs to look like and then we talked a bit about the on the demand supply side how do we look at representing the digital twin and trying to take all of the information in terms of inputs collaboration your process capabilities and trying to create these optimized plans in terms of supply chain financial outlooks for the plans etc so that that would be the other one and then in snop what we're thinking like well like i mentioned moving to these online meetings right so from the static powerpoints where you're hard to get information now using nlp and digital meetings and example of this would be let me just go here so in our platform what we're thinking is online meetings where i can create new presentations and assemble what a meeting needs to look like right so i can go into a meeting i can actually add commentary to the meetings and run these meetings in real time and also collaborate during the meetings right so i can create my snop meeting decks here online and then use this to actually collaborate review the meetings have feedback from various roles that are part of the meetings and so on so i would be an example of what we're saying in terms of snop moving from static powerpoint to live on the system then the other art of the possible that we are thinking about is uh really taking from snlp to true ibp right so so far we do snop meetings almost clear on course game right so they do a little bit of post game to understand what happened and we look at it as the three w's right what happened what is likely to happen and what actions to take and then what happens you need a lot of post game analysis and root pausing what's likely to happen is where you bring in your analytics to do predictive analytics your automated forecasting and then your scenario planning capabilities into what is what prescriptive actions to take right and this is where the gap closure workflows are extremely critical because when you have a demand upside for example how do i convert that into a supply response right and those are things that uh we want to incorporate in the tool itself right so for example here let me just give you an example uh if i were to submit a forecast then how do i make sure if there's a demand upside then what supply side scenarios do you evaluate right and what objective functions to maximize so then there's some automation to it it automatically runs these scenarios and comes back with the best response scenario as well so trying to pull in some of these in terms of when out of the possible is what we're looking at uh with that let me just skip through so there's some more details on some of the observed challenges that we see we've covered a few of these but i'll skip that same thing on the supply planning side on the snop side as well and then users are important because they're different types of users i think what we wanted to talk about executives your planners and analysts operations it innovation teams and so on and each of them has a need and what we're trying to connect is uh is bring them all together rather than them having their own applications all of them need to be interfacing with the planning systems to provide their feedback as well so with that i know we're up on 15 minutes i'd like to leave some time for q a but let me uh in the meantime just quickly touch up on these value potential and talk about the framework that we have how do we derive value from some of these uh applications right and implementations typically what we see is in terms of the value buckets of how value is spread in in planning and integrated planning and operations is mainly around cost reduction right so we have these four categories reduction in cogs reduction in inventory improvement in productivity and improvement in sales right so those are the four or five four buckets that we typically see and within them we've broken it down further so there's improvement reduction in your manufacturing cost itself based cost based transportation or logistics cost reduction and expediting and flex cost obsolescence material costs etc so those are really the the buckets in terms of how we look at value and typically what we've seen is for a billion dollar company it's about uh 10 to 20 million dollars in in value that we can save through implementing better digital transformation and better implementation of ibp systems right and the details behind this is here uh which is really breaking it down in terms of if i were to look at uh transportation costs then how does it help improve right through early visibility to constraints so you're actually looking ahead and uh and and reducing your peaks in peak periods and trying to optimize the flow and a great example here is a walmart right what we had to do there was uh sort of smoothen their flow because they were having a lot of situations where you have orders for halloween or orders for christmas coming during halloween time periods right and so their entire network was bottlenecked and trying to at least smooth that out understand when the orders need to flow in and smoothing it out those are some of the capabilities of the digital uh twin and the supply planning that we do but these are details behind what are the elements and drivers that we look at and how do we add 09 convert this into value buckets so that as we go to these customer engagements we're able to create uh if you focus on on increasing sales then what are the key drivers to to help the businesses right so with that i know we're up on time for q a so let me skip through we've talked about a couple of customer cases but there are some other cases as well here which we can cover it or you guys can read the deck if needed but there are some examples on our website as well uh for you to go through and understand what the challenges were what was the project timelines what were the key value drivers and why online was selected for some of these implementations and we do have some benefits as well in terms of again trying to give you a breadth of these various solutions across industrial manufacturing you know this is steel manufacturing tire manufacturing etc and i would like to then maybe call it the wrap here i'll end with a couple of slides here uh just to get you guys present um and get involved right so we have what we call the aim 10x network and innovators network which is really aimed by aim we mean ai powered management systems for 10x improvements in in planning and decision making and this has a collection of folks that that are part of it so it'll be good if you want to uh mix with uh with industry peers or mix with like-minded folks then this is a great place for you to start to network with some of the best people talk to people if you want mentorship opportunities or if you want to discuss with like-minded folks and just to give you an example of some of the folks you look you'll be talking to folks that are former execs at png uh avon kraft heinz etc so we have a number of people that are part of the network now and if you are interested you can go to our website we have the aim 10x innovators network you can look it up on linkedin as well and look at joining that okay so with that let me go to the last slide where we can open it up for q a and maybe use it to wrap up the discussions as well okay a good uh this is but i'd like to thank very much you message for this very thorough presentation on the art of possible mostly from a supply chain uh integrated business planning perspective now uh one one of the reasons for having uh umeshin online come here to present is to highlight a disruptive uh new style companies like 09 uh are bringing new and new innovations to to the ecosystem that have that can have a very deep influence on the the capabilities of many players uh in industry and also to see how such companies evolve and are growing a portfolio of of offers and capabilities so at this stage what i'd like is to entertain uh if we have any questions or comments from the audience uh we'll we'll we'll get there uh so we have a question from uh shubham uh that says can you please uh give uh provide a real-time use case of collaboration between tier 2 tier 1 production facility and distribution in the manufacturing space so obviously this would take a lot of time but if you could like in one or two minutes just give the spirit of that uh this is done that would be appreciated i can do that so let me actually show it on the tool itself so it'll give you a better flavor for what we're talking about so typically when we say extended network right so i didn't show you the i showed you the knowledge graph right where we presented um all this stuff here so when we talk about the knowledge graph itself you can go all the way from what i showed you the suppliers to the supplier suppliers so you can model not just your plant locations but your suppliers plant locations and this can be extended right so i have suppliers i can have my tier one tier two manufacturing again you could have a deeper network but what we want to do is also pull the collaboration within the tool itself so i may have contract manufacturing collaboration where if i have suppliers where i can share with them in real time right i'm showing them what is my production needs and then i can get feedback from them where they can come in input their comments around can it be supported or not pull them together similarly if i have supplier collaboration where i go to tier one or two to supplier where i'm sharing with them what is my procurement forecast and what are their supplier commits collaborate on the supplier commits in the tool so they are actually able to login to the tool to actually provide their capacity and give us an information about is it supportable or not or do they have capacity restrictions based on that we can run a constrained plan right and then we can figure out what is the information that they have provided and use that information to to actually help us guide the the decision making process right so rather than having unconstrained plans all the time we are actually working tier 1 tutor suppliers can log into the same system to actually provide their capacity or their issues that they're having and use that to actually collaborate on the orders if there's any constraints in in their specific plant locations then use that information to input it in the tool and reflect that in the in their commits and use that to replan the the supply itself uh well i have a few questions that are related to uh essentially the the kind of time frame and engagement related to uh to the kinds of solutions your your innovations bring to the market so how much right now of the companies the clients that you you guys are dealing with uh are using your a system more like to solve a problem and your staff is mostly driving it versus them taking ownership of the of this and making it their own and related to this is what kind of timetable are we talking about from starting to think about this and having the kinds of capabilities you've been this deep uh displaying through the presentation i think we run into these various companies at different levels of maturity there are cases where we've gone in and phase one for example if you do this incremental and one of the pieces that i skipped over was how do we move these projects to incremental value right and that was the slide let me just switch back here which is this one how should enterprise approach the transformation from silo to this digital brain and this is where i think the fast and iterative so a lot of us are working in the agile model now but instead of these uh multi-year projects what we're doing is just incremental capability what are the quick fitters in terms of white spaces where we can quickly prototype and start deploying those right and what we've seen at various customers is just taking that approach starts showing them the value of these solutions for example let me see if i have some numbers here wilbur alice for example it took them about 10 weeks to go live into production for demand planning right and so we're what we're trying to see is the short projects rather than year-long projects where they're looking at small incremental solutions and that helps them really get familiarity and build the trust with the systems for example tata steel was within five months they were live right on on the system itself for uh for supply planning so that will give you an idea of uh not long projects but short projects that are very agile where we're working with the customers to actually build the solution and present some of these reference models that have out of the box capabilities but being able to extend it and and that's the piece where we try to come in with the initial implementation we do the initial pass but the intent is that it's a self-service uh platform and i haven't shown you the self-service capabilities but uh we end up training the it teams and their project team so then they can start continuing to configure and extend the platform on their own uh once they're familiar with how to do that so we don't need to be involved anymore it becomes more of a self-service for them once they're applied what i'll do now is kind of blitz the reserve questions okay you answer like in 30 seconds so that we can wrap up this with uh with just a twist at the end uh you talked about simulation what kind of simulation are you guys really doing in this is this discrete event is just like sensitive analysis what kind of simulations are you currently doing you can run your own experiment design of experiments where you can set up simulations in terms of various input parameters are different or my demand is different run various scenarios to come up with the recommendations so it's a combination i would say of manual user inputs with discrete events and design of experiments and there's automation as well i can run certain scenarios automatically in the back end every time i want to run the scenarios okay now what what is your perception about deterministic heuristics or or decision making versus more stochastic approach that will invent the stochasticity into the heuristics or decision process we have been somewhat deterministic in the sense that it is it is a heuristic for sure we do have alternate optimization plugins as well where the framework allows us to put in lp solvers and so on but those become extremely slow and hard to explain so our solver is mostly heuristic but it is deterministic in the sense most of the demand that comes in from customers is deterministic but what we're moving towards is more of a stochastic the demand should not be determined again that is something that we're working towards to convince the customers and there's a maturity journey there but we would like to move in that direction in terms of stochasticity of the demand okay now in terms of uh pushing innovation even further uh have you guys already explored uh integrating into your overall technology and capabilities uh leveraging virtual reality that one we haven't uh we are looking at partners in that respect we've been starting to work with microsoft on some of their uh the surface glasses and so on where they have some of these uh prototypes already in place and we are trying to come up with maybe joint collaboration opportunities where some of these digital collaboration that we're trying to do in the platform can be easily integrated with vr tools and they're able to leverage both of them coexisting because we are cloud-based we are in azure we are in google cloud so we have some capabilities where we are collaborating with microsoft on and hopefully in the near term if there's a good use case then we will try to integrate okay good so but two two nice things about this is that once you're you're still even though you you've already achieved great innovation threads here uh essentially you're still open to two more and also that you're thinking collaborative ways to make this happen knowing that each company has boundaries there and i like through the process how much you interact with the with users with your clients to to shape okay what what your technology is all about so uh on this we have to close this uh this nice uh meeting i appreciate very much your openness and the thoroughness with which you you've you've presented and then answered questions uh you have this coordinates okay uh um sorry from online solutions he's executive vice president for product management uh at online uh and please um please uh go with me in in thanking him and feel free to connect with him afterward as you see we'll have a few more uh seminars that are interacting uh with with online this semester and uh hopefully you'll be able to to make it uh for for these as they are personal to you so thank you very much everybody for attending i thank you especially 242 umash for this great job and we'll see you for next uh seminars the next one on february 24th take care bye you

Channel: GTSCL

Views: 412

Rating: 5 out of 5

Keywords:

Id: YhHhQQJL1t8

Length: 86min 53sec (5213 seconds)

Published: Tue Feb 09 2021