The Infosys global supply chain management blog enables leaner supply chains through process and IT related interventions. Discuss the latest trends and solutions across the supply chain management landscape.

« April 2011 | Main | June 2011 »

May 31, 2011

Is there a need for Next Generation Supply Chain planning products?

It is no major news that organizations are consciously gearing up to capture a customer demand regardless of the source/channel of demand and then attempting to fulfill the same. Internet dot com sites, cell phones/ ipads or any other such mobile devices give the buyers the necessary platforms to capture and convey data from multiple points and attempting to move to a truly smart consumer-centric model.

While there is a lot of activity and discussions on the trends in the Next Generation Commerce solutions and how a multi-channel e-commerce (MCC) solution enables retail organizations to offer consistent services across virtually any customer touch point, the question that comes to mind is "Are Supply Chain Planning systems (both demand and fulfillment) in a state where they can predict and plan the increasingly volatile demand?"

While having the ability to capture a firm demand, irrespective of the channel undoubtedly is a great solution and is helping a lot of retailers in increasing their top lines, would it not be a better situation to be in, if one can anticipate buying patterns across channels & plan for this volatility in demand well in advance?

Retailers are currently riding high, reaping the benefits of using the MCC solutions and having the ability to meet or exceed sales targets. It would be a matter of time before, surpassing the newly set sales targets become the need of the hour. There are already discussions around MCC being a thought of the past and an emerging era of Agile Commerce. However both the situations, i.e. a matured MCC market and the Agile Commerce era seem to prompt for a better planning system; one that can predict the volatility of demand at multiple channels or sense and predict the demand at multiple customer touch points.

While the buying patterns in  B2B scenarios are relatively more stable,  buying patterns in B2C scenario's could have a lot of changes from one period to the next. To the best of my knowledge, forecasting solutions that are currently available in the market from well known ERP vendors or even the Best of Breed demand planning solutions are either based on Statistical, Regression based or consensus based models. However most of them do not carry the models for demand sensing and cannot predict how many customers would be visiting the kiosks, or how many would be visiting the websites and how many would be visiting the stores? More importantly, of all such visits at multiple channels, how many of such prospects would actually buy at each of these channels. 

The primary input for the current forecasting solutions for demand planning is past or historical sales accompanied with firm independent and dependent demand inputs. While these can continue to be the prime inputs, the forecasting algorithms can aid a lot more if they can be changed to include the buying patterns of customers to predict better.

Larger businesses like Unilever and P&G have identified this gap and have implemented additional demand sensing solutions like those from Terra technologies to improve their responsiveness to current sales and/or reduce their forecast accuracy error over and above their demand planning solutions. 

In this era, where the critical success factor of  businesses lies not only in its ability to forecast and plan, but also being able to respond quickly to changes, it would make a lot of sense even for the software product companies to have demand sensing go hand in hand with demand planning and thus evolve the Next Generation of Supply Chain Planning products.

It would be interesting to get more insights & opinions on this. Do share if you have come across situations where you have seen these requirements or know about best of breed vendors or packages that are offering solutions in this direction.

May 30, 2011

HazMat in SCM Needs to Worry About New Tech Impositions!

During the last couple of weeks, I have been associated with the brainstorming around how best to manage Hazardous Waste or Material (HazMat) for a major US retailer. The client management team is focused on having a solution that covers HazMat of all kinds across the enterprise since that is the core KPI of the department. Most of our initial discussions have been around two threads:
1. Understanding our Point of View on Reverse Logistics (RL) since HazMat typically need to flow in the opposite direction of the regular product supply chain flow
2. Figuring out whether SAP EHS solution is the best bet versus IBM Sterling Commerce RL capability and IBM Maximo's work management capability.
Personally, I wasn't too keen on going for either of these approaches.

For me, the fundamental question was neither RL as a functional domain nor the three competing product offerings in question, but that of who are the people using HazMat and when/where do they use it. Broadly, this meant classifying HazMat occurrences into two buckets:
1. HazMat coming out of core supply chain operations at various stages triggered by product returns via multiple combinations like (a) returning entity - customer returns or vendor returns (b) disposition type - product returned via backhaul or handled via one of the authorized disposition mechanisms for that SKU (c) Damages during receipt in the inbound PO-processing cycle at the warehouses.
2. Hazardous consumables being used during Facilities Management (FM) activities. The work orders being executed as part of a typical FM task can involve something like a corrosive floor cleaner (acid) which would need to be tracked as part of consumables request pertaining to that specific work order.

If I place the user group or department up in the front, then what this means is that the customer service representative accepting a return or maintenance personnel cleaning that messy floor already has a system in which they do their day-to-day operations. This is where I would advice any HazMat information to be captured as this means lesser change management burden, reduced user angst, more accurate data being collected and consequently, improved tracking and audit of HazMat. Having a one-size-fits-all department-driven design launching a new system/application could well mean the opposite of all these.

The first phase is currently being planned for Retail items, which makes it more RL-oriented than consumables used during FM tasks. That means

  • enhancing existing OMS and WMS functions to capture HazMat
  • processing blind/non-blind orders better, with better linkages to original sales order
  • having more relevant and detailed disposition codes
  • implementing alert management functionality triggering management personnel about the presence of such products returned/damaged
  • product vendors or 3rd party providers taking the material away from premises subscribing to these process flows and so on.

If the systems can be built to prompt the right levels of safety measures to be taken as the transaction happens, that would be of real discernible value to the frontline employee handling the HazMat. If the reporting engine can be enhanced to classify, slice/dice HazMat data in terms of SKUs, damages vs returns, return to vendor vs disposition managed by the retailer and so on, that would keep the compliance hawks happy as well.

Net-net, we need a thought process that weaves in HazMat as part of the regular supply chain (or maintenance) processes, helping the handlers use the same systems they were using and then building in supply chain analytics (near real-time and data warehouse driven) at the back-end around those transactions that had some presence of HazMat in them. Treating HazMat as a product attribute which triggers certain behavior would result in a much simpler supply chain with all the required capabilities than giving it a life all of its own.

May 28, 2011

Infosys @ IBM Software Days, Dubai

I am just back from attending the IBM event - Software Days at Dubai.

Infosys was one of the GOLD sponsors of the Event. Primarily meant to show case the Smarter Planet initiative, the event had a reasonably good participation from customers and partner community. Apart from the 8-9 partner's stalls, IBM had put up separate stalls to show-case their capabilities in various verticals like Government sectors (I found the smart city story from Rio de Janeiro quiet interesting- too good to believe! This can potentially take E-governance to the next level), Retail sector (Sterling Commerce et al) and Natural resources.   

The forenoon session was dedicated to sessions by IBM leadership team in the region. The one which caught my attention the most was the session by Middle East & Africa General Manager Takreem El- Tohamy. Lots of insight into evolution of IBM into what is it today over 100 years. (I didn't know that IBM had such a significant role in Apollo 13 mission. The video on this was quiet informative.)

One of the main attractions for me in the event was the keynote address by former world chess champion, Gary Kasparov- my childhood hero. In my experience, best of the chess-players generally are not best of orators. Found Gary to be an exception. He spoke on the topic- 'How IT changed the way human think' based on his famous Man vs. Machine encounters with Deep Blue in the mid-90's. He played with his machine opponent, who could analyze 200 million positions in one sec, whereas, as human being, he can at best think of 3 positions in 1 sec! His opponent has, over past several years, deeply analyzed and internalized all of his past games, and is well prepared. Whereas he had never seen his opponents play in the past. Quiet an unfavorable situation to be in- even if you are world champion for 20 years! Still he managed to win. The reason, according to him, is that human can think of 'WHY', whereas machine can only talk about WHAT, WHEN & HOW. Machine cannot shift priorities in between the game based on WHY - the man can. That's where human has a huge advantage over machine. He went on to suggest that the best and ideal, unbeatable combination would be a Man + Machine. Human will do the strategic thinking part and machine will be there to prevent him from making any possible 'blunder' moves.

Post lunch session was divided into 5 parallel tracks- on collaboration, BPM/SOA, Application life cycle management, Business Analytics and one on Business Infrastructure and Services. Since I had a specific interest in asset management in utility space, I choose to attend the session on Infrastructure and services. I particularly liked the talk by Jim Fletcher; chief architect of Tivoli Solutions on Facility Management for Smart Building. He lined up quite a few statistics to support the need for energy management in Buildings (I am not sure to what extend the audience - mostly from oil-rich GCC countries could relate to need for energy conservation!). He explained how well the newly acquired TRIRIGA for space management fits into the Energy Management Solutions- supported by Maximo in the Operations Management.

Overall, a well-managed event show-casing IBM's commitment for the region. Would have been better, if there was more on-stage participation from partners and customers like what they managed in IBM PULSE event last year. Hope to see this happening in the years to come.

Subscribe to this blog's feed

Follow us on

Blogger Profiles

Infosys on Twitter