It is known as the second largest moon in the solar system. Closer home, Titan is a company that manufactures over nine million watches. But the similarity doesn’t end with the name.
With its trademark signature tune and a brand name Titan has ruled the consumer durables industry for over two decades — much like the race of powerful deities by the same name that ruled during the Golden Age in Greek Mythology.
Established in 1984 as a joint venture between the Tata Group and the Tamil Nadu Industrial Development Corporation, the company today has a customer base of over 80 million.
Its watches are sold through six different market channels including — an international market, 240 World of Titan showrooms, over a hundred Time Zone outlets and more than 12,000 retail outlets.
On the other side of the supply chain, out of its 800 vendors, 70 percent are key and are responsible for over 80 percent of Titan’s inventory. These include stainless steel, tool steel, leaded brass, engineering plastics, components, etcetera to make the specialty movement and finished watches.
All its market channels are serviced by 32 CFAs (clearing and forwarding agents). The network up to the CFA level is the integrated supply chain group, which is responsible for making right products available at the right time, at the right locations, within constraints of costs, location and capital.
But what exactly is ‘right’? How does the company know which market demands for which specific type of watch, which manufacturing unit should cater to which retailer? And exactly how many watches will he need?
The answer, rather feebly, lay in an inaccurate demand forecasting system.
The system was not dependable as it hinged on the opinions of the field sales force — opinions that were not backed by numbers for the most part.
A Little Too Late
Ironically, the company that sells time to the world, found itself lagging behind. The way out was an effective ERP system. And that is why Titan implemented SAP R/3 in 1999, to replace its legacy system. This took care of basic issues like integrating all transactions on one platform and ensuring better market alignment. It also checked rising inventory and corresponding costs.
But, there were still some issues this system overlooked. Accuracy was virtually non-existent. Titan also needed a decision support system to support the company’s supply chain. Titan’s OLTP (On Line Transaction Processing) system was unable to handle the 2,200 variants (variation in terms of presentation style of the same product) of finished goods.
They needed a system that would — even with expansion — produce a forecast that would be instrumental in directing the company through a minefield of capacity constraints, multiple sales geographies and a multi-tier distribution channel.
“Co-ordinating between the supply chain and distribution has been one of the biggest challenges of Titan. So, in 2003-2004, we implemented almost every possible module of ERP. What we still needed was a good forecasting system, which could predict sales across the country, given the complexity of new models, promotions, and other constraints,” says N. Kailasanathan, VP and CIO, Titan.
On the supply side, staff would work out a supply forecast based on current stock, and likely sales over the next two months, and then plan for production.
“All of this was done on manual worksheets, which were sent to the factory where weekly feeds were given for production plan. There was no scientific basis for any of this. Plus, the entire forecast process took 15 days,” says J. Murali, group manager, supply chain and logistics at Titan.
This paved the way for another tight spot: mismatched inventory. The production alignment ratio hovered around 70 percent — which meant that on an average 30 percent of Titan’s inventory was not being used.
“We asked our regional co-ordinators for estimated regional demands. These demand figures, based on gut-feelings, went to the central all India co-ordinator. Since we operate on 90 days of production-dispatch lead-time, this process was unable to give us uniform figures. Ultimately, actual sales against forecasts would be as low as 40 percent,” says Murali.
A Stitch in Time
The company was beginning to feel the pinch. It realized that the ever-expanding world of Titan could get difficult to handle if processes stayed where they were: archaic.
Battling unreliable forecasts and mismatched inventories, among other things, Titan felt that it needed an application that could take care of demand planning, production planning, procurement scheduling, shop floor scheduling and distribution planning.
A tall order, but Titan knew that it was already running on borrowed time.
So the IT team, comprising core experts from production, supply-chain and sales pulled up it socks and started the groundwork on evaluating which solution would work best. After studying other companies with similar needs, they made a short list. “We got experienced consultants to tell us where our present supply chain stood and how we could optimize it,” says Kailasanathan.
Titan zeroed in on an Advance Planning System (APS) which could deliver forecasts that took into account resource constraints like plant and supply capacity. The study threw up three options: SCM 2.0 from i2, a solution from Adaxa and SAP’s APO solution (Advance Planner and Optimizer). Titan decided to go with APO for functional reasons. “SAP gave us more advantages because we already had a stabilized SAP ERP backend,” says Kailasanathan.
The project kicked off on April 2, 2004. In its first phase, the objective was to integrate supply chain design and implement the demand planning module. The second phase would see the implementation of the supply network-planning module and in the third phase, the production planning and detailed scheduling module.
Present Tense
All said and done, the implementation was no less than a Herculean task. It was plagued by more problems than one. Getting data from 200 retail stores and about 100 redistribution stockists on time was a challenge. Apart from that, they were impeded by hardware limitations. “The system would crash frequently when we ran the optimization model. That was one of the biggest pain-points and it delayed our go live date,” says C.S. Ramesh, GM-IT, Titan.
Optimizing the SNP (Supply Network Planning) put Titan in a quandary. “We had to break the network into logical splits so that it could run sequentially. This went on for four to five months. We wanted to first streamline the processes in SNP — including production planning. But establishing the optimizer was a nightmare.”
To tame that bull, the vendor suggested that Titan move up from 32bit to 64bit. “They ran it on their server, using our database, and only when they confirmed the stability of the 64bit did we start using it,” Murali recalls. Titan went online with the new 64-bit system in June 2007, and now they have worked up a usage capacity of up to 48GB. There is scope for future expansion, says Kailasanathan, because currently usage is at about 16GB to 17GB.
But slowly, the primary reason for deploying the APS was fading: checking forecast errors. “The system forecast error level was very high, running into unimaginable numbers like between 150 and 500. SAP told us that this was because we had excessive data, which was not conducive for good quality forecasting,” says Murali.
But all was not lost but there was light at the end of the tunnel.
“APS’ are not like ordinary implementations. They have fewer users — most with a stake in the business. So they themselves became anchors and started using the system. It took some persuasion to get users to accept the optimizer figures, but once we showed them how accurate they were, everyone came around,” says Ramesh.
That Moment in Time
Good times were just round the corner. The project brought with it a bundle of perks for Titan. The business benefits came, largely, from demand and supply planning — both of which were illogical earlier. The new system provided a streamlined demand planning process.
Importantly, the system allowed Titan to create semi-finished watches (in the first stages of production most watches look the same) and wait for a confirmed forecast to produce finished goods. This strategy of postponement improved Titan’s market response time and ensured that they made fewer ‘non-sellable’ watches.
The system also took advantage of the deployment optimizer, which allowed Titan to distribute finished products from its five assembly plants on an optimal basis. The Dehradun and Baddi (Himachal Pradesh) plants optimize their supply-planning schedule — they have been designed to ensure that constraints in availability, distance and demand are met.
The data received from the forecasting system is run past field sales personnel who provide their own insights to the figures. “We run the forecast at gender, price and brand level. And forecasts are segregated to variant level and that is being used by sales forecasting,” says Murali. The implementation brought relief from high product alignment ratios. “Alignment improved significantly — from 70 percent to about 85 percent,” says Kailasanathan. “This in turn reduced the stock of finished goods, in relative terms.” It also allowed Titan fine tune the mix of the products it had in the market.
The system also significantly reduced production planning time from a week to three days.
As far as tangible benefits are concerned, Kailasanathan points out that since the project was deployed, production figures have gone up. “Since we went live, production has increased between 40 percent and 45 percent over a three-year period. From 70 lakh watches, we are now at 100 lakh. We have also opened more outlets — 170 to 240,” he says.
The number of manufacturing units has also increased from two to five, and additional complexities that have cropped up have been effectively taken care of by the new system — without any additional resources. Expansion, Kailasanathan maintains, is more of a strategy decision as technologies grow. And this is facilitated when processes are streamlined, making operations smoother.
Future Perfect
‘Nobody knows what the future holds’ — a phrase Titan is alien to, simply because it has its eyes set on tomorrow. In the end of 2007, Titan moved to the next version of APO — the SCM 5.0. The next step will involve a deployment of assortment planning on the APO. “This is a specific module in the APO and we plan to exploit it for our World of Titan showroom needs. It involves forecasting the right product mix required for different showrooms depending on an optimal mix of brands, gender, color, etcetera,” Ramesh says.
They also plan to use a function called ‘real time deployment.’ This function will alert Titan to re-stock an outlet once stock levels fall below a fixed level. “It is expected to go live any day now,” Ramesh adds proudly.
From letting time call the shots to knowing what the future holds, Titan Industries has empowered itself to move ahead of time. The melodious signature tune plays on.
