This research represents the first assessment of dynamic time-based postponement. The study should be replicated in different research settings such as the apparel industry, or seasonal products such as candies or sports equioment.
The apparel industry presents many of ihe same characteristics as a limited-time offer, but the number of orders that can be placed to manufacturers Is limited and lead times could be as long as nine months, Frequenily, retailers can place only one order to a manufacturer for the whole season. This kina of problem is known as the newsboy problem. Dynamic time-based postponement could be used to complement the analysis of the newsboy problem to determine the impact to the whole supply chain of increasing the number of orders that can be placed to manufacturers.
For example, two scenarios could be modeled. One would represent the actual practice in which only one order is placed for the whole season. This scenario
232
should include the expectation of sales at regular price, clearance sales, sales through secondary channels and iosi sales due to the inability io adapt decisions to follow demand more closely. The other scenario would include the same factors, but it should incorporate the fact that decisions could be revised. This means that the original order could be adjusted within certain limits. This analysis would result in the size of the business opporiuniy to be shared among ihe members of the supply chain who made this new scenario possibie.
seasonal producis such as candies and sport equipment are similar to fashion oroducits, but candy prices are lower and replenishment cycle times are snorter than for the apparel industry. The value of sport equipment products is higher than candies and apparel, and replenishment cycie times tend to be shorter than in the apparel industry. in the sports equipment industry, orders can be placed more frequently than in the apparel industry, but not as frequenily as in the candy business.
High product value and the ability to place more than one order might allow management to identify several lead time options which suggests that dynamic time-based postponement could nave potential. Testing dynamic time-based postponement in several business setting will provide a better understanding of the implementation issues.
An optimization mocdelrepresents a goal for which management would sirive.
In this research, this goal was potential cost savings fo the supply chain. It was a goal because the costs savings are based on the output of a model andi in the model all activity are performed exactly as planned. That is, there is no contingencies, no human errors, and no miscommunication or iate communications between managers. As a next step, a simulation-based model could be developed using the output of the optimization as the target inventory policies. This analysis might provide insights in terrns of the dynamics in the supply chain. A simulation model can be designed to replicate the managers’ decision-making processes and would result in a closer representation of an actual LTO environment. For
233
example, despite the inventory levels set by the optimization before the LTO starts, if managers observe that demand is lower {or higher} than expected, they might not strictly follow the plans. Instead, they might adapt decisions to their new expectations. A simulation model could be developed to capture this dynamic nature of decision-making.
The develooment of a case study of an actual implementation of dynamic time-based postponement would provide insights œaboult the intercompany relationships ana the management concerns that hinder the development of supply chain-wide intiatives. The potential contributions of the develooment of a case study includes: the identification of specific management concerns; how these concerns are presented to the other members of the supply chain, now these concerns are received, addressed and solved; how risks and rewards are determined ana shared among firms; and, the impact of the experience on the degree of closeness of the relationships between firms.
Another future research opportunity is fo reassess past studies in postponement by changing the sequence of activities, and to identify each of the benefits of postponement oresented in Figure 2.9. Figure 2.9 shows that postponement allows demand aggregation, reducing the forecasting horizon, and the learning effect. These three sources of benefits from postponement ullimately provide cast savings. The benefits of each might depend on environmental factors.
This research opportunity has potential to result in a prescriptive framework fo assist management in the determination of the benefits from postoonement. lf the majority of the benefits from postponement stem from the learning effect, then the priority should be to develop the capability of information sharing and the infrastructure fo support information visibility. If most of the benefits stem from reducing ine forecasting horizon, then ithe focus should be on improving forecasting techniques. If the benefits stem primarily from the demand aggregation effect, then efforts should be concentrated on commonality of parts and subcomponents.
234