- How about measuring the cost of forecast error status quo?
- In real $ terms. How do I know which forecast model to use?
- How do I know what proportion of demand ‘happened’ because of promotional event?
- What do I lose by forecasting in weekly buckets instead of monthly?
- What do I gain out of it? Do I? Should I have centralized demand planning or regional demand planning?
- How does competition do it?
- What does ‘best-in-class’ do?
- Should I forecast at SKU level or brand level?
- What do I lose or gain in doing so? What really is (should be) the purpose of forecasting?
- How about new products?
Your customers do NOT care how good or bad is your forecast error. They bother about delivery service level. A reliable predictable service level. But the good thing is it doesn’t cost much to improve the forecast accuracy that is only one of the determinants of service level. Yet the benefits of forecast accuracy are counter intuitive and highly non-linear in nature. Studies say that a mere 5 % improvement in forecast accuracy and staying good at that can potentially reduce inventory investments by order of magnitude upto 35% or more over an year. Without compromising on the service levels.
Apart from deep subject matter expertise on model building and demand side analytics, The Management Technician brings diverse insights on demand planning practices, processes and organizations from a variety of industry contexts. We work on diverse applications like SAP APO, SAS, i2, Demantra and other best of breed applications to make the solution deliver superior results. Our custom model tuning and macro services have objectively benefitted our customers by achieve higher forecast accuracy over time and more importantly a feeling of trust in the solution. Something we excel at.
Supply and Capacity Planning
It takes quite some labor and prep work to figure out the right assumptions even to fulfill distribution demand, let alone customer demand. Esp. with multiple sources of supply and demand (warehouses, customer warehouses). Even in a medium scale supply network with 5 locations, 50 SKUs, 100 vendors and 1000 customers, supply planning can get daunting. Supply chain dynamics can quickly get complex. But your systems perhaps does not live the reality right now. Products are growing. Customers are growing. Some products are being discontinued. Sometimes with conditional restrictions. There are external conditions imposed by Customers, Sales, shippers, vendors and even the law. All leading to generation of a sub-optimal supply and distribution plans if these facts are not considered real time. A plan that not only fails to meet the demand at the right time but is also expensive by virtue of inventory investments. Based on obsolete assumptions. Despite multi-million dollar systems in place you struggle to balance inventories across locations, the most basic of activities!
We have helped customers re-think new paradigms in planning. Since we understand the corruption by variability at work. Inaccurate lead times, over realistic safety stocks, source of supply restrictions, resource and material availability constraints have a crippling effect on your ability to service demand. Much more than what you possibly assume. Despite a lot of capacity and a lot of inventories on hand, the stuff isn’t there in the right place at the right time. No, it isn’t just about your ability to ship. It is about your ability to meet or exceed stated service level by deploying the inventories at the right place at the right time. That needs a synchronized response to demand. It is your ability to simulate a variety of supply plans before you go ahead with production and procurement. It is your ability to ‘model’ your supply and demand ‘nodes, contracts and relationships correctly before you ‘run’ your supply and replenishment plans. It is your ability to distinguish between source of demand (customer demand, distribution demand, forecast) and source of apply clearly at all times.
Deployment and Fulfilment
Keeping promises is hard. Even when you have all the material and resources at your disposal. It just isn’t there the minute it was needed. Execution may be a ‘no-brainer’ but then try fulfilling a multi-product order with products made at 4 locations and you will quickly realize 80% of your orders were never completely filled. You either rationed some bit here and there. Changed your source of supply. Substituted some alternate products and pushed something else that your customer didn’t want. But these aren’t the only constraints.There are physical constraints too. No trucks. No space. Customers have logistical restrictions. Then there is all sort of ‘custom’ requirements expected from ‘standard’ products. Your order to ship processing costs could have been much more expensive than you thought. But thankfully there are tools like Global ATP we creatively set up to manage plethora of distinction that needs to be made in customer priorities and product shortages.
Integrated Business Planning.
Coming soon from our partners
Supply Chain Analytics
Analytics really isn’t an isolated thing. For any supply chain analytics to be useful, it needs to be business as usual. Use the insights from diverse info sources to make crucial day to day transaction decisions. While some Analytics is about post facto metrics that may be useful for quarterly appraisals and performance reviews, the Analytics that matter the most in your planning and fulfilment functions are things like responding to volatility in demand or taking necessary actions to minimize cost over-runs or even decisions like whether to go ahead with producing a whole batch of 3000 units to fulfil an average monthly demand of 200! That means macros and metrics that you must have as you walk. What was the price elasticity of the current promotion? How should I revise my forecast? What is the marginal cost of making this batch now? What is the probability of stock out for this product in next 10 days and what must I do now? What is the profitability of this order if I only partially fulfil item#5 at a discount of 50%? May be more profitable than doing 100%!