لطفا منتظر باشید ...
A Hybrid Model
One last look at the concept of manufacturing planning and control (MPC) and its component master production scheduling reveals some interesting implications and opportunities for SCM. Three concepts integral to MPC that may be subject to enhancements by the JIT philosophy are freezing, time fencing and consuming the forecast.Freezing is a production planning practice whereby no changes can be made to the production plan (and hence exploded MRP requirements) inside of a specified period. In many organizations, it is not unusual to see a frozen period in the range of 8 to 12 weeks. For international organizations that do not have a lot of confidence in their supply chains, the frozen period can be much longer. This exercise has implications for MRP due to the fact that raw materials inventories are built up in anticipation of upcoming production, regardless of what the real demand turns out to be.Time fencing is a continuation of freezing in that measurement periods are broken down by weeks, with changes to the production plan only allowed outside of a certain "fence." Consuming the forecast, conversely, is a practice that implies replacing forecasted amounts with actual orders. As sales orders come in for weeks outside of the frozen period, adjustments are made up or down to the forecast based on the collective volume of all orders received for those periods.One major difference between MRP planning and JIT is that the former defines periods in weeks and the latter in days. It is the expression of demand in days that allows JIT practitioners to place orders with suppliers literally on the same day production will take place (if, of course, the supplier is close by). Another difference is that JIT normally does not have frozen periods, at least not for weeks at a time. This practice delays or postpones the buildup of raw materials inventories to the benefit of the buyer's balance sheet. Finally, JIT is supposed to be based on actual demand, so consuming the forecast should not be a consideration (certainly for periods closest to production). Given this scenario, how do supply chain managers build greater flexibility into their models while assuring product availability?A possible solution to this dilemma is to continue to utilize the MRP component of the MPC model, but express demand in days rather than weeks. This not only allows for greater precision in planning but also creates a situation where the level production of all products can be achieved. Also, if companies can attain the dual goals of reducing cumulative lead times and keeping them stable, the possibility exists to reduce the frozen time fence and as a result the buildup of inventories.Finally, if it is not feasible to execute orders with suppliers on a daily basis, consumption of the forecast could be a viable alternative. Use of this concept in conjunction with a reduced frozen period allows for greater exactitude in determining raw materials requirements, with purchases based on actual demand rather than forecast. Although not perfect, this option does offer the potential of greater accuracy in aligning inventories with order-based requirements.Without question, the linchpin to the above or any other supply chain exercise is the accuracy of the forecast. If projections were accurate to begin with, there would be no need to carry inventories or safety stocks. Unfortunately, the human condition is an imperfect vessel brimming with uncertainty, and as long as people are completing forecasts (aided by technology or otherwise), there will be errors. There is hope, however, based on the length of the forecast period itself.An axiom of forecasting is that the shorter the period under projection, the greater the probability for accuracy. A simple example illustrates this point. Most people know with certainty what they will be doing tomorrow. It is much more difficult, however, to predict what one will be doing in six months (compensating for bad forecasts, no doubt!). Understanding that forecasts correspond to planning periods and that the length of the period is a function of the longest lead time for a given product or component, it is fair to say that if lead time can be reduced, so can the forecast period.Again, the shorter the forecast period, the greater the probability of accuracy. This concept involves logic and mathematical modeling that are outside the scope of this discussion. Suffice it to say, however, that lead time management and its impact on inventory levels once again makes its way to the forefront of SCM. These ideas, when combined with the possible solutions presented for reducing min/max levels, postponing order points and diminishing the need for safety stock, make a powerful argument for the improvement of supply chain performance.
