“Welfare organization Self-sufficiency” – Supriya Kummamuru
I’m Supriya and this paper was co-authored by Ravi who is not here. He could not travel and Mr [MGPL 00:00:09] whom I think you’ve all interacted, and myself. I’ll take about 15 minutes to present so I’ll make up for the time and ensure you’re not late for lunch. The purpose of writing this paper was to actually showcase one of the case studies or projects that we have done. We are a consultancy organization and most of our work actually relates to consulting in the IT domain. We are known more for IT and technology services, but we also have projects where we do consulting and mostly in the domestic market. The purpose here is to showcase a model which we have developed. It is a proprietary to [Tata 00:00:55] consultancy services, and it was actually based on concepts of cybernetics, and it has evolved over several applications which we have done in the management consulting area.
It’s two-fold. One is to present the model which we have, and the other is to show you how it was applied in the context of a consulting assignment. Like you said this is two-fold, one is the name of the model. The way we have said it’s called cybernetics [influence 00:01:26] diagram. It’s based on Norbert Wiener cybernetic concepts which facilitate capturing aspects of complex systems and help in problem resolution. That’s we have … If you see he’s talked of its application across different disciplines and I think we have applied it to our business domain. Second is to present the case of complex tribal welfare organization where the problem context was understood using this model. We could convince the client where the solution lay with coming off with the model of the context that they were looking at. The presentation outline basically is I’ll showcase the model and then give you the case application.
I will not dwell much on the theory because I think we’re all are familiar with the philosophy which has been spoken by several speakers during this conference. How the model was evolved, I’ll spend a little time on the model itself. What you can infer from it, and then we’ll go on to the case. Norbert Wiener on cybernetics, this is what we all know. That it’s a theory of messages, I has been applied across different disciplines, and here we’re using this discipline to apply to the domain of management. We’re all familiar with what it means, the Greek word, and how we use the concept of communication, control, and feedback in order to ensure the system steers towards its purpose.
In the consulting scenario why did we use it? We’re all aware that as we move where the world is getting more connected, I think complexity is increasing because it’s difficult to predict the behavior. Today’s guest speaker today morning said it’s human aspect even in any autonomous system it’s difficult to predict the behavior. You can first order cybernetic talks of control communication and you’re able to predict. When we come to second order, where you have the observer and humans as part of [these 00:03:43] systems, it is difficult to predict of the behavior, come up with solutions. We see it in everyday context, so we were thinking of what tools to apply and then most of the answers that have come up in different domains, they need a lot of data where you actually quantify. If you go back to system dynamics there are models but they’re all based on data being available. We work in a context and scenario where our data is very not available, difficult to obtain. We need to come up with answers in very short times.
In this scenario what is it that we can model, and what kind of answers that we need to come up, was the challenge. We actually came up with this qualitative model which threw up answers in a short time, and too a large extent they were optimal and correct to the client context. The approach here was in the consulting world you need to to have a speedy and approximate solution. It needs to be in qualitative terms and you need to have the finer details, and once you come up with the answer maybe you apply a specific model to come to mole quantification and give the specifics. This was the approach that was taken, and this is the framework that broadly tells you how it is done. These are the steps. When it was evolving there were no defined steps because it was done through application through several consulting assignments, and we looked at … It was a trial and error which is the best[to it 00:05:28] and over almost 20 to 24 years of consulting practice.
When the center was started it was called the systems engineering and cybernetic center. The connection to Norbert Wiener is Mr FC Kohli, who’s actually known as father of Indian IT, was a student of Norbert Wiener and unfortunately he could not travel to Boston. He was keen on travelling but the doctor didn’t permit. He’s now 92 years old. He was a student and he started the center in 1982, and he brought in [Preceipi and Murthy 00:06:04] who headed this group and was working in the area of systems thinking and cybernetics. He had applied these concepts, came up with several models applied to consulting. This is one of the models which we arrived at. This is the framework that we go through, I will not through the steps, this is just to show you these are the steps and the methodology that we evolved.
We’re a consulting practice. Basically there are four stages at the end of which you come up with something called a cybernetic influence diagram. Cybernetic because of the character of any context that you look at. There is communication, there is feedback, and there is control. You have both human and non-human interface in that. We had yesterday’s speakers talking about it, and out of it we were trying to derive what is it that you can assess out of it? There are three things that we, at this point, actually infer from these. One is we look at loops which may be there in the system. Then we look at two points which we call the margin base points. I’ll come to what those mean.
First let us look at what a feedback loop is, and I think most of you in this room know what feedback is. We have been talking about it since yesterday. There are two kinds of feedback in any system, it could be a positive feedback or a negative feedback, and not necessarily the sense of a negative connotation. Positive feedbacks are those which reinforce an action, and a negative feedback is one which controls an action. Here you have one where I’ve shown you a sale cycle where over time if you’re limited population whom you’re directing your reach, then the potential customer goes down. Whereas in the case of simple [bank 00:07:58] balance and interest earned, it goes on reinforcing so the amount goes up. The other is in a context you have multi-cycle structures, where you have a collection of several feedback loops. Positives and negatives, when you try to actually analyse the context, and what we need to do is you need is you need to see which are those important [routes 00:08:19]? Which are those clusters that you’re looking for in a context and try to intervene.
The intervention points out of this model are two. Basically [the 00:08:30] influenced and the influencers. Let us look at the influenced. It’s simple to understand from the this representation. It’s a point where lot of arrows actually in pinch , which means that these are points which get influenced easily. Then you need to see what is that that you need to do to bring about a change in that? It requires a lot of effort to bring about change here. Simple illustration, if I go back to [Phil’s 00:08:58] presentation, if you’re talking of sustainability. How do you ensure it … There is so many factors which you need to act on to actually have a sustainable environment. It’s not just those three, probably goes beyond those three elements that he talked of.
The other point which you can identify from this context is an influencer. Where you have a point out of which a lot of arrows emerge. It is from the diagram that you get. This is a point which you can actually use to leverage the context. You can actually intervene, give suggestions, and probably come up with answers to it. In a consulting or any context that you’re trying to understand, if you can try and intervene at a [bus 00:09:43] point you’ll be able to come up with a solution. How do you leverage the model? The model allows us to quickly sift through information and represent our understanding of customers in a systemic perspective. It presents information in a form that is easily explainable and facilitates further inputs from stakeholders, subject matter experts. It helps to check for completeness, presence or absence of factors, thereby refines our problems. Helps validate the understanding of the client system.
This is [a 00:10:20] model, I’ll show you the application where I’ll actually show you how the output looks like. This is the case illustration of organization which was looking at helping the tribals in ensuring their viability. This organization was set up to help the tribals in a forest area in a part of an Indian state. They had two activities, one was to help them sell the produce which is made in the forest, which is non-timber produce. By helping them process it, transport, and sell it to retail outside. The other was, since they live in a forest, they had to look at the basic requirements where they had to provide them with food supplies procured from the outside market. They were limited to that area but this agency was supposed to support it, which is [a 00:11:17] government agency. Currently the problem they came up with was they were not a viable organization. It had to be heavily funded by the government, because they felt these two activities was leading to huge losses for the government, and then they had this intervention when they wanted to look at what exactly was wrong with it.
That’s when Tata consultancy services was approached, and this is what we did to understand the problem. What we did is we first discovered the context, I’ll tell you in short what we did. We applied the model, came up with the whole context in terms of the cybernetic influence diagram, where we captured aspects typically representing what is there in that particular system. Then if you see … I’ll come to the next diagram. We diagnosed what was wrong, if you see … Physically you can see there’s separation of two entities in this. What it meant was there was a commerce and welfare aspect to it. Commerce was where they were trying to sell the produce made by the tribals to the outside world. The welfare context was where they had to bring in produce from outside and provide it to the tribals at a minimum cost.
If you see the agency was mixing up both the issues, where one hand they had to actually ensure the price that they bought for the welfare activities was [talked 00:12:44] and it was at a lower cost. The commerce activity was, there was a notion among … It clearly showed there was separation of concerns, there was separation of entities which were involved with each of these context, but still when you spoke to people there, stakeholders, there was a mix up on that. This clearly brought out that difference between these two activities. With the help of a market research agency there were some notions which were dispelled in terms of pricing, and using this approach we separated the two concerns. The activities of the two contexts were separated, and solutions were given in terms of design. Where the welfare activities of the organization had to be separated. Where the credit activities and procurement activities has to be separated, and the commercial activities. Where the value addition had to be done and pricing had to be linked to the market.
Based on these two there was the cash flow statement generated which eventually led to them coming out of the cash crunch they were in. The attempt here was to show how the model was used to actually showcase the different activities, the entities that were involved. Then coming up with the design intervention which helped the tribal corporation to come out of the red that they were in. This was the design. At this point I stop, so if there are any questions I can take. There’s a lot more information which [I’ve 00:14:22] not shared since the purpose was only to showcase the model, its application, and the win that we got out of it, so …
Very good. Okay, thanks Supriya.