Lead time in 3 minutes

Lead time can be summarised as the latency between the initiation and completion of a process. In a supply chain, this is usually involved whenever goods are purchased, transformed or serviced.

While this may appear pretty simplistic and not worthy of further investigation, we would argue that this is not the case. Thinking about lead times is every bit as important as thinking about future demand. The time, normally measured in days, between what you decide to do today and the time it takes for the impact of that decision to be felt, actually shapes how the entire supply chain operates.

The importance can be clearly seen if you ask yourself the question: “Why do we need to forecast at all?” Forecasting stock requirements allows you to prepare for the delay of stock being delivered and helps to avoid potential stockouts. In order to do this, the demand planner needs to anticipate how much inventory will be consumed between now and the next replenishment - whilst taking into account the fact that, whilst those goods are in transit, inventory is gradually getting depleted at the same time.

As such, lead times represent a lower bound of the maximal agility that a company can achieve. As a rule of thumb, a company remains committed to its past decisions for roughly the duration of this period. Logically, this means that the longer the lead times, the higher the risk if market conditions brutally change.

So where do the key challenges lie? Often, when we have little or no experience on lead times, we rely on what our suppliers promise us. This value frequently goes unchallenged in IT systems - sometimes for years - and a single fixed value doesn’t take into account the wide variability we often can experience with our lead times. This is also a downside of the safety stock model, which we have discussed previously.

Typically, the answer to this is to provide a buffer to account for this instability. But this approach usually falls short. Firstly, because lead times generally follow seasonal patterns, such as shipping delays around holiday periods like Chinese New Year. Secondly, because if a delay occurs, it often can be significant. For example, your supplier may experience a raw material delivery issue or custom control issues.

To conclude, this why we, at Lokad, take a probabilistic approach to lead times. We strive to understand the various patterns, rather then assuming and relying on averages. The key message is that predicting lead time is hugely challenging. Whilst often this can be somewhat controlled by a “smart” purchaser, there are many incertitudes which can only be managed by adequate tools. As such, lead times should be looked at as the counterpart of demand and dealt with accordingly.

Find out more about Lead Time.