Managerial Implication

From the above scenarios and sensitivity analysis, we can give some advices to the express company.

4.5.1 The simple recovery process

From the small network case, we find that the recovery activities are taken in following sequence:

1) Change to the alternative paths 2) Re-allocate own trucks

3) Rent trucks or flight containers

4) Rent the different modes (train or ship)

First, the path will change and then the own trucks will be re-allocated. Only when above changes can still not satisfy the demand, the model will rent the additional modes. Finally the different modes will be chosen, for example the train or ship.

Figure 4.21 The simple recovery process

Recall that the goal of our research was to find the possible recovery activities which can

(Figure 4.21) according to the aforementioned sequence in the small network case. The process can tell the express company the rough direction in the pre-incident stage. When the disruption occurs, the company identifies the location, affected type and severe level of disruption first. Then, find the similar scenario. They can refer its recovery activities and revise the actions by this process. Should we change to the other paths? Should we re-allocate own trucks? If there is no similar scenario, we also can directly evaluate each recovery

activities step by step by experience. With more severe situation, the company needs to consider more recovery activities together. This rough response action only can use in the short term. If company wants to consider the recovery activities in the medium-term or long term, it needs to execute our resilient model to get more precise quantitative solution.

4.5.2. The suggestion for the reactive recovery actions

1. Mode choice

(1) Sea transportation is advantageous in the short distance to be the substitute for air transport during the disaster, for example , Keelung port to Shanghai Seaport.

(2) If rail transportation wants to substitute for the trucks which service between service centers and airports, it depends on the transportation and transshipping time. If sum of them are shorter than the transportation time for trucks, rail may be chosen during the disaster, for example, rail in Japan in our scenario.

(3) In our large-scale network case, rail is a nice substitute for air in the continents of Europe and Asia, for example, Shanghai - Moscow - Europe.

2. Route choice

(1) From the scenario and the sensitivities analysis, we know that objective functions are not sensitive in transportation time in the certain range. Only when it is over a threshold, it affects the objective functions. Thus, instead of choosing the route with shortest shipping time, the express company can choose any routes which allow the cargos be delivered before the cargo time value declines. We infer that it is because of the assumption of our cargo time value function. If there are different type of functions, the rule maybe different.

(2) The sensitivity analysis shows that the low renting cost is more significant than the

transportation time to impact the objective functions. Renting time and re-allocating time have smallest impact on them. Thus, the express company may consider the renting cost first then the transportation time in the certain range because the time becomes less sensitive due to our cargo time value function.

3. Carrier selection

(1)In the scenario of Japan earth quake and failed Hong Kong airport, we find most of trucks are rented by carrier 1, because its renting cost is cheaper than carrier 2. But the renting time of carrier 1 is much longer. It shows that the time spending on renting and dispatching the trucks (RT) by different carriers is not the significant factor to impact the carrier selection.

From the sensitivity analysis, we also can find the renting cost (RC) has more impact on the objective function than the renting time (RT). Thus, we consider that the key effected factor for carrier selection is renting cost (RC).

(2) In the scenario of failed Hong Kong airport, we find that when the rented capacities are few, it may rent the trucks from carrier 2 to save the time spending instead of saving cost.

According to above finding, we suggest that when the rental capacities decrease, the carrier selection can change from low cost carrier to the carrier with short renting time.

4. Cargos arrangement

From scenario analysis, we find technologic products (short-life cycle commodities with high values) are always transported with the entire amounts and on time. The holiday gifts will be considered in the second order. The constant value products are most easily abandoned or transported with long travel time. Thus, sorting the cargos with different type of time value function and transporting them efficiently can help company to reach higher cargo time values.

For the express company, we are going to suggest them to consolidate their resource to handle the cargos with high value or cargos with highly time sensitive and transport different types of cargos with different routes. The way can make express company reach the great efficiency to get higher customers satisfaction during the disruption.

5. Renting activities vs. Trucks re-allocation

(1) From sensitivity analysis, we find no matter in renting activities or truck re-allocation, the cost parameters effect the second objective function more significantly. So the express

company can pay more attention on the cost saving rather than time saving when it makes the decision on the recovery activities.

(2) On the other hand, the second objective (the incremental resilient cost) is more sensitive in the renting cost (RC) than the re-allocating cost (VC). Thus, we can infer that renting activity plays more important role in recovery than the truck re-allocation. To the express company, saving the renting cost (RC) is more efficient than re-allocating cost (VC).

(3) Own truck re-allocating activity is the smaller recovery. It changes with route choice, so company can re-allocate the trucks by experience if the disruption is not complex. But when it is complex, the company has better to recover based on our model’s solution.

From the above suggestions, we find the renting cost seems to play the important role on the recovery. Time becomes less sensitive during the disruption. It is because the cargo time value function we assume. Due to the time value function, the cargo only needs to be sent before the cargo time value declines. Thus, in the certain range (not over the time threshold), the model can do the more flexible decision on time and the main affected factor is renting cost in this certain range.

4.5.3. The suggestion for the proactive resilient strategies

Although our model provides the quantitative tool which is used after the disruption, we also can use it in advance to help express company draft proactive strategies. For example, we can give the express company some proactive suggestions based on our large-scale scenario analysis.

1. Prepare the backup hub

From the large-scale results, we find having the backup bub in the network is important. When the critical global hub like Hong Kong Airport in DHL failed, the shipments will be consolidated in the Shanghai Pudong Airport (SHA) instead. If there is no backup hub, the recovery cost should be very high so that the company is prone not to recover. Thus, the express company is necessary to invest in the redundant global

transshipping hub before the disruption. It can lead to a better recovery.

2. Develop the mainly alternative route depending on our results

We also can suggest company to develop the alternative route before the disruption happens based on the result of model, for example SHA- LEJ or HKG- DBX-LEJ in our big network scenarios. The company can maintain few flights in the alternative routes in normal times or ensure the availability of rental capacities in the alternative routes. Once the disruption occurs, the company can use these routes quickly.

3. Sign the different type of contracts with carriers

From sensitivity analysis, we know that when the own capacities are few, the company is going to rent more capacities from others. However, when there are few available renting capacities(AK), the company can’t recover the cargo flows. Thus, it should focus on the resource acquirement after the disruption. We suggest the express company can sign the different type of contracts with carriers.

(1) Choose specific carriers which have the specific routes to be long term contractual partners (long term partnership)

(2) Sign the emergency contracts with non-long term partners (short term partnership)

In the scenarios of Icelandic volcano and Japan earthquake, most of aircraft containers are rented from non-partners because the contractual partner doesn’t

accommodate the capacities for the requisite routes. The same situation also happens to the sea transportation. Thus, the selection for the air and sea carriers is limited by the

According to the above results, the express company can use our model to do several scenarios analysis. These scenarios may have big impact on the company or are with the high probability. Based on the result, the express company can choose specific carriers to be long term contractual partners. For the carriers who are not the long term partners, we put stress on signing the emergency contract with them to ensure the timely acquirement of capacities.

4. Appropriate reserved capacities

From sensitivity analysis, increasing more own capacities only can reduce a little resilient cost and increase a little cargo time value. The benefit of increasing own capacities is getting less. It shows a diminishing curve. Thus, maintaining appropriate reserved capacity is a better choice to company.

Because the proactive recovery decision problem is not in our research scope, we don’t know how much capacities the express company should maintain before the disaster. So we give the suggestions to the future study that providing the model which can consider both proactive and reactive resilient decisions.

On the other hand, it is worth noting that by choosing either resilient strategy or activity, express company will have higher cost. There is the question from these companies. Is it worth to spend such great amount money to recover the transportation activities? As far as the long-term business goal is concerned, the added cost is kind of saving the customers’

satisfaction and commercial goodwill. By the increasing disruption events, the companies can’t be passive role to do nothing anymore. They need to consider the benefits of their customers and create the good enterprise image so it is necessary to spending on the recovery activities. From the other standpoint, we also realize that the cost plays the key role in the express company. Thus, in order to use the less expenditure to reach the great effects, our model considers two objectives simultaneously which is maximizing total cargo time value and minimizing the incremental cost from recovery.