| Introduction | | | | The traditional approach is to calculate a moving |
| Over the past 10 years a great number of | | | | safety stock based upon a number of weeks |
| businesses ranging from SME’s to Blue Chips | | | | forecast, sometimes “refined” by using |
| have been rolling out or updating their Enterprise | | | | forecast variability against historical usage (in APS |
| Resource Planning (ERP), Distribution Requirements | | | | systems). This method is fatally flawed in 2 ways; it |
| Planning (DRP) and Advanced Supply Chain Planning | | | | relies exclusively on forecasts in order to calculate |
| solutions (APS). There are a variety of reasons for | | | | the amount of safety stock required, and it actively |
| this intense activity, ranging from the need to | | | | plans in a level of ‘dead’ stock, with the |
| consolidate IT following an acquisition, through to the | | | | anticipated on-hand levels moving between the |
| desire to improve the IT capability in order to | | | | safety stock level and safety stock plus the minimum |
| implement a particular supply chain strategy. | | | | order qty. |
| The 10 key elements for a Best Practice Supply chain | | | | The new approach to inventory target calculation |
| 1. Have a clearly understood and agreed service level | | | | sets a maximum target level of inventory for each |
| agreement (SLA) with your customers | | | | SKU. This is made up of an element of inventory for |
| The SLA should be a detailed understanding of the | | | | the replenishment time plus an element for demand |
| service to be offered, particularly in relation to lead | | | | variability, which is statistically related to the required |
| time, minimum order qty and stock holding | | | | service level from the stocked item. This approach |
| requirements. It should also articulate the parameters | | | | makes the entire inventory available for use, with |
| that define exceptional demand (e.g. a promotion) | | | | on-hand levels fluctuating between the inventory |
| from normal fluctuations in demand that can be | | | | target and zero. It also builds in some sound statistical |
| accommodated as “business as usual”. | | | | probability of material availability based on historical |
| 2. There should be a robust, regular channel of | | | | demand variability. As long as your S&OP process |
| communication with your customer, in order to | | | | flags up demand that falls outside of this agreed |
| measure and improve performance levels defined in | | | | variability, you will have a lean level of stock that |
| the SLA. | | | | supports your customer SLA at least cost. |
| Most enlightened businesses now have some kind of | | | | 7. Completely separate Planning activity from |
| Sales & Operations Planning (S&OP) processes. Many | | | | Execution activity |
| however are very inwardly focussed and don’t | | | | Another curse of MRP is its ability to blur the line |
| include sufficient or any direct input from the | | | | between planning and execution. A planner is being |
| customer. This is the opportunity for the customer | | | | asked to replan and chase orders daily or even hourly |
| to communicate significant future demand changes | | | | as the MRP “shuffles the order pack” each |
| for which the supply chain needs to be re-calibrated. | | | | time it runs, requiring their constant attention. It is |
| 3. Proper supply chain planning must consider total | | | | vital to separate the activity associated with planning |
| business cost including Demand, Capacity, Supply & | | | | from that of daily order raising and execution. Best |
| Inventory Planning. | | | | practise requires that a plan be set, normally for a |
| Another common failing of many S&OP processes is | | | | month, in line with the frequency of the S&OP or |
| that they do not cover all the elements of cost. | | | | forecasting cycle, and then execution happens daily |
| Typically the debate can be around manufacturing | | | | against this plan, enabling a set of lower skilled or |
| efficiency and capacity and ignore the costs | | | | automated actions to be taken daily. This normally |
| associated with poor customer service or resultant | | | | means a key change to the skills required by a |
| inventory. A good S&OP process understands the | | | | planner, meaning considerably fewer but more highly |
| service model agreed and then determines the least | | | | skilled individuals. |
| cost way of delivering this. | | | | 8. Execution tools that allow orders to be raised in |
| 4. Know when and when not to use a Forecast | | | | line with appropriate replenishment rule. |
| Forecasts, no matter how inaccurate, are the best | | | | There will invariably be the need to cater for a range |
| tool that we have to determine future capacity | | | | of replenishment rules when placing manufacturing or |
| requirements. Therefore we should have a toolset | | | | purchase orders, from fixed repeating schedules, |
| “A common mistake here is to confuse Demand | | | | through Kanbans and re-order cycle items, spares |
| Variability calculated entirely from the historical | | | | requirements, to pure Make to Order. Most ERP/DRP |
| demand pattern with Forecast Variability, which is the | | | | systems support some but not all of the required |
| variance between history and forecast. The former is | | | | techniques. Therefore you will either need a new |
| correct the later is meaningless”that enables us | | | | order generation tool that uses the required |
| to easily access this information. Forecasts are | | | | execution technique to compliment the chosen |
| typically not bad at determining how much of | | | | replenishment rule, or you will need to imaginatively |
| something we will need, i.e. is demand increasing or | | | | configure your ERP/DRP systems to behave and |
| decreasing, but very poor a predicting exactly when | | | | raise orders differently. |
| the demand will occur. Therefore never use a | | | | 9. Forecasts must be completely eliminated from the |
| forecast for order generation, to do so flys in the | | | | ordering/execution process.“There is a belief that |
| face of any Demand Driven Lean approach. | | | | by extending lead-times you increase your available |
| 5. Segment SKUs based upon their demand volume | | | | capacity. This is a myth” |
| and variability and then select the appropriate | | | | “Inaccurate Forecasts are the major cause of |
| replenishment rule for each segment | | | | cost in all supply chains, and Forecasts are always |
| The same service level and/or replenishment rule is | | | | inaccurate!” |
| rarely appropriate for all SKUs. Normally there is a | | | | Inaccurate Forecasts are the major cause of cost in |
| range of items from high volume, low variability items | | | | all supply chains, and Forecasts are always inaccurate! |
| that require a highly repetitive supply plan, through to | | | | The aim should be to never execute an order against |
| those with sporadic requirements that should ideally | | | | a Forecast. Forecasts can however be used as an |
| be 'make to order'. This segmentation fits closely | | | | indicator of forward demand volumes and linked |
| with the principles of Lean Manufacturing. A common | | | | correctly with actual demand variability from history |
| mistake here is to confuse Demand Variability | | | | can be used to set appropriate inventory policies and |
| calculated entirely from the historical demand pattern | | | | targets. |
| with Forecast Variability, which is the variance | | | | 10. When planning, use the shortest possible planning |
| between history and forecast. The former is correct | | | | horizon in order to minimize the likelihood of plan |
| the later is meaningless. | | | | change and to minimize the number of orders that |
| 6. Use the correct replenishment rule to calculate the | | | | need to be controlled. |
| correct stock level for each SKU level, to satisfy the | | | | There is a belief that by extending lead-times you |
| agreed customer service level in the SLA. | | | | increase your available capacity. This is a myth. |
| Once all the levers of cost are understood and the | | | | Extending lead-times will upset your customers, |
| appropriate replenishment rule selected for each SKU, | | | | particularly if this violates an agreed SLA, and |
| an inventory and production plan can be built that | | | | increase the level of activity and cost required to |
| delivers the desired customer service Levels. In order | | | | plan. Wherever possible, drive down lead-time, which |
| to get the correct balance of inventory and | | | | will in turn drive cost out of the supply chain. This |
| manufacturing cost, a new way of calculating | | | | approach fits exactly with the requirements of lean |
| inventory holding is required, that flys in the face of | | | | manufacturing. |
| much of the conventional inventory planning wisdom. | | | | |