This cool button delivers CIO stories to you on Facebook:
How do you support India’s largest steel manufacturer, with revenues of over Rs 15,000 crore (2011 figures), with an IT team of just 17?
A) Superman is masquerading as a CIO.
B) The 17 IT staff are actually robots with no ‘stop’ button.
C) Its IT system is extremely resilient.
Clearly, it will be a long time before options A and B are possible. As light-hearted as the question is, it throws hard light on the game-changing amount of flexibility Essar Steel has built into its systems, a feat unparalleled in its space.
If there’s one thing Essar Steel knows well—apart from steel, that is—it’s how to push the envelope. It’s, for example, the first steel company to set up an end-user distribution chain for steel products. It’s the world’s largest steel retail chain with a network of over 375 retail outlets. It also owns India’s largest steel processing and distribution facilities with an annual capacity of 4 million tons.
That retail strategy, however, took a big bite out of the company’s efficiency. On any given day, Essar receives a lot of specific, custom orders—across a wide range of products, from automobile body parts and ship building and from boilers to heavy engineering structures. For example, a girder, one of its many products, could have multiple variants depending on chemical composition, physical attributes or surface treatment.
“Every material has to be defined in terms of many characteristics which can be as high as 25 and needing a quality plan which needs derivation of at least 250 characteristics. Most orders are specific, and require a high degree of process customization,” says Suneel Aradhye, CIO, Essar Steel.
In 2006, when orders hit 35,000-a-month levels, each customized order was still being processed manually. That took a toll given the complexity of steel manufacturing; raw material is subjugated to numerous specific treatments for a certain chemical composition, mechanical properties, and dimensional tolerances.
The way Essar dealt with the problem was to attach a code, called material code, to each new product it created, a list that got longer by the day. In 2006, the number of material codes in the company’s SAP system had already hit 100,000.
Because it was so complex—and the costs of being wrong were high—sales staff lacked the confidence to complete orders by themselves. Orders were finalized only when a staffer with metallurgical knowledge, preferably someone from the quality department, was involved in decision-making. That slowed the decision making.
“Due to the confusing three-way interaction between sales, quality, and production planning, it could take up to three to four weeks to book and finalize orders,” says Aradhye.
Inside the Crucible
Things only got more complex when the world’s only vertically-integrated steel company (it has businesses from mining to retailing) disclosed plans to increase its production capacity by 40 percent within a year. Aradhye saw a thin line drawn between Superman and a deep abyss—and he was walking on it.
But as luck would have it, Essar was considering a SAP re-implementation at about the same time. “We were running on an older version of SAP, which was due for an update. I decided to ask the guys at SAP if they had a solution to my problem,” says Aradhye. SAP consultants advised Aradhye to look at SAP’s Variant Configuration (VC) tool.
Thus began the course of evaluating Essar’s internal processes. “Earlier, SAP ordering happened when a salesperson input market demand into the system. Orders were then processed by the quality department over an offline database with appropriate characteristics and metallurgical knowledge. The two databases were not in sync,” says Aradhye. That was a problem the VC tool could help fix.
Implementing the VC tool meant that all variants had to be reflected across every integrated system involved in processing an order. This was akin to converting the system’s metallurgical brain into rules decipherable by machines. A small error meant tons of incorrect products and a threat to the plant’s physical wellness.
The dice was loaded against Aradhye. Essar isn’t the only Indian business that deals with extremely customized orders, but few had attempted to use the VC tool. Worse, the reference cases he came across seemed to have ended in failure, a fact his management brought up. To top it all, he wanted to make changes to the core of a running business. “We met with serious apprehension from the business when we proposed changes to process design principles. It called for separate efforts on change management. References to failed efforts in the past were also a challenge in our bid to win the confidence of the business community,” he says.
Undeterred, Aradhye forged on. During the initial implementation, variants were calculated and evaluated (in Excel) for each process and sub-process. While the plant’s capability based on inputs from quality and production planning were mapped, Aradhye’s team also ran checks against 10-year-old order book to build a robust prototype. Business plans in terms of plans to manufacture new grades of steel for next two to three years were also considered.
It was a lot of work, but, in the end, worth it. Today, the VC tool has become an indispensable part of Essar Steel’s daily operations. It trimmed material master data from 90,000 to about 260 configurable materials belonging to 450 different products. “If we had not implemented the tool, by now Essar’s system would have generated 500,000 unique material codes—a picture of chaos!” says Minesh Mohile, GM and chief solution architect-IT at Essar Steel.
By simplifying the system, Aradhye also shrunk opportunity-loss by cutting sales order processing time by a factor of seven. “Today, the sales team makes a data entry, prints out the contract, gets it signed, and finalizes order sitting right across the client. The whole order process is completed in three days flat,” says Aradhye.
There are other benefits as well. The project helps improve customer service by achieving the pre-agreed time lines and lowering cycle time. It also curtailed data entry errors and saved about 3,000 man-hours a year in supply chain management.
Sometimes, all it takes to be different is steely resolve.