Nepal Earthquake Disaster Response Supply Chain — 2015

Earthquakes are not uncommon in the foothills of the Himalaya Mountains. But nobody can predict exactly when they will happen, so each one is a surprise. And then people have to respond quickly. They need to: figure out where the earthquake hit; get damage and casualty reports; and do something useful to help. The screenshot below shows the supply chain set up to deliver aid supplies to Kathmandu, Nepal after an earthquake. Aid supplies arrive from three global depots where aid supplies have been stockpiled to be available for emergencies such as this.

nepal-sc-high-level
(click screenshot to see larger image)

This scenario is modeled on actual events that occurred in April and May of 2015. Supplies are flown in from depots in: Brindisi, Italy (upper left); Shanghai, China (right side); and Kuala Lumpur (lower right). The depot at Kuala Lumpur also sends supplies by ship to a freight depot in Kolkata, India (lower center). At Kolkata the supplies are offloaded from ships and transferred to trucks for delivery to Kathmandu.

Structure of the Nepal Disaster Response Supply Chain

A distribution center (DC) is set up next to the Kathmandu airport and local warehouses are established in the areas of Nepal that suffered the greatest damage and casualties. In the the first week after the earthquake large amounts of supplies are delivered by air freight and by truck to the Kathmandu airport DC.

The screenshot below shows country level detail for the disaster response supply chain in Nepal. The icons used for facilities in the supply chain are standard icons used by the United Nations and other disaster response and humanitarian aid organizations.

nepal-sc-detail(click screenshot to see larger image)

Kathmandu airport DC is the focus of deliveries for aid supplies from outside the country. Supplies are stored in the DC, and from there are moved by truck and helicopter to the local warehouses. There are four local warehouses. They are: Dhading Warehouse (upper left); Babare Warehouse (upper right); Charikot Warehouse (center right); and Deurali Warehouse (lower right).

Helicopters and trucks are used to deliver supplies to the warehouses in Dhading because the road to that location is narrow and the going is slow, so trucks alone cannot meet the demand for supplies. The warehouse at Babare is supplied only by helicopters because it is located up in the mountains and there are no roads to Babare that trucks can use. Trucks deliver supplies to the other two warehouses because the roads are adequate (load this supply chain model into your edit screen and zoom in on the four warehouses; switch to satellite view and see the actual locations and the condition of the roads).

Modeling and Simulating the Disaster Response Supply Chain

This supply chain is modeled by combinations of four types of entities placed on the map: 1. PRODUCTS; 2. FACILITIES; 3. VEHICLES; and 4. ROUTES. Examples of these entities from the model are shown below. Data used to define the entities in this case study was provided by the World Food Program Global Logistics Cluster (http://www.logcluster.org/). See more about the Global Logistics Cluster in this short YouTube video – https://www.youtube.com/watch?v=DvNJx8F9j4k

The screenshot below shows  a group of the four entities that moves products between the Kathmandu Airport DC and the Deurali local warehouse:

  1. Size and weight of a PRODUCT called “Food (pallet)”
  2. Storage capacity, internal product demand, and product on-hand amounts for FACILITY called “Kathmandu Airport DC” — production of food at the depot represents food that is acquired locally
  3. Cargo volume and weight, and speed of a VEHICLE called “Medium Trucks -20” which is composed of 20 medium trucks
  4. Round trip time and distance on ROUTE called “Kathmandu – Deurali” taken by those 20 trucks to deliver products from Kathmandu to Deurali, plus amounts of different products they drop off at Deurali — route is shown in blue on the map

nepal-sc-entities
(click on screenshot to see larger image)

This model assumes other vehicles at the local warehouses move supplies to individual villages and local aid stations in the vicinity of each of the warehouses. The daily product demand numbers shown for a local warehouse represent the combined demand from all the villages and aid stations served by each warehouse. Data for any or all of the entities can be changed as needed if better information becomes available.

NOTE: To further explore the logistics operation within the vicinity of a local warehouse, a lower level supply chain model can be built. It would begin with a local warehouse and show the vehicles and routes that move those supplies to villages and aid stations supported by this warehouse. The warehouse facility would produce amounts of products equal to the amounts delivered to it as shown in this higher level model. Vehicles such as SUVs, mules or even human porters moving on routes that may be little more than dirt paths can be accurately simulated in this lower level model.

[ Register on SCM Globe to gain access to this and all other case studies. Click the blue “Register” button on the home page (www.scmglobe.com) and buy a subscription (if you haven’t already) using a credit card or PayPal account. Then go to the SCM Globe library and click the “Import” button next to this case study. Scan the “Getting Started” section (if you haven’t already), and you are ready to go. To share your changes and improvements to this model with other SCM Globe users see “Download and Share Supply Chain Models” ]

FIRST CHALLENGE: Get supply chain to run for 15 – 30 days

Import the “Nepal Disaster Response Scenario” supply chain model from the online library and open it up in your edit screen to inspect the model. Click on the menu tabs for Products, Facilities, Vehicles, and Routes. Select entities in those menus and look at their data. Zoom in on some of the facilities and switch to the map view to study their location and the surrounding area.

Then click on the “Simulation” button and run a simulation to see how well this supply chain works. You see trucks and planes and ships running on their routes, and real time displays on the right side of the screen showing day to day inventory levels for different supplies.

As the simulation is running click on some of those data displays for the supply depots and the Kathmandu airport DC and local warehouses to see the trends in the on-hand amounts of different supplies. You see various product inventories are trending up and others are trending down at different facilities. Then on the fifth day (as shown below) the Kathmandu airport DC runs out of construction material.

nepal-sim3
(click on screenshot for larger image)

The question now is how to fix this problem. You can increase the amount of construction material shipped into Kathmandu from outside sources. Or you can look at the distribution and consumption of construction materials at the facilities in Nepal and see if there are opportunities to fix this problem by better managing supplies within Nepal.

Understanding the Flow of Products through this Supply Chain

The simulation results screenshot above shows delivery routes for supplies flowing from the Kathmandu Airport DC out to the four local warehouses in the areas that had the greatest damage. The diagram below is composed of the on-hand inventory graphs for the Kathmandu Airport DC and the four warehouses.

nepal-supply-flow(click on diagram for larger image)

This diagram shows inventory levels trending downward at Kathmandu, but inventory levels trending upward at the local warehouses indicates supplies being sent are exceeding demand at those warehouses. It looks like a good way to fix the running out of construction products and other products at Kathmandu Airport DC is to reduce product amounts being sent out to the satellite warehouses.

The diagram below is composed of the on-hand inventory graphs for the Kathmandu Airport DC and the four depots outside Nepal that sent aid supplies to Kathmandu. It shows inventory levels at the four depots to be holding generally stable with a few exceptions. In the Shanghai Depot and the Kuala Lumpur Depot the green line (food) and the purple line (health supplies) are trending downward indicating need for adjustments there.

The “V” shaped pattern of on-hand inventory at the Kolkata Freight Depot shows the drawdown of inventory as it was shipped to Kathmandu, and then inventory levels rise again when the ocean freight shipment arrived from Kuala Lumpur.

nepal-supply-flow2 (click on diagram for larger image)

Mental Models for Managing Supply Chain Operations

From scanning these on-hand inventory graphs one can see the patterns and trends of inventory as supplies flow into Kathmandu, and from Kathmandu to the local warehouses. It looks like reduction of the flow of inventory from Kathmandu to the warehouses will fix the problem of declining inventory levels at Kathmandu. There may be no need to increase shipments into Kathmandu.

Go ahead and make those adjustments. Edit the supply chain model and reduce the drop off amounts for products on the routes from Kathmandu to the four warehouses. Or reduce the number of deliveries by increasing the delay between departures for the vehicles that run on the delivery routes. Then run some simulations and see what happens and what works best.

As you make adjustments to the model and run simulations, you will develop a mental model of how this supply chain works. You will get an intuitive sense of how adjustments to any of the four entities (products, facilities, vehicles, and routes) affects the other entities. After you solve the first problem, other problems will arise. There are many possible solutions to the problems you encounter, and the problems you encounter will depend on the actions you take to solve previous problems.

Keep trying different adjustments until you get the supply chain to run for 15 – 30 days. There is no point in getting the simulation to run for more than 30 days because after 30 days the situation in Nepal will certainly have changed, and you will need to change the supply chain design as well.

Once  you get this supply chain to run without problems for 15 – 30 days, then come back and work on ways to lower operating costs and inventory levels at the facilities in the supply chain. The mental models you develop for this supply chain will help you see opportunities to lower costs by better use of vehicles and transportation options, and see opportunities to lower inventory levels by better matching of product deliveries with product needs at each facility. The intuitive sense you develop in this case study will also serve you well in the real world to help you manage similar supply chains.

SECOND CHALLENGE: Combine Collaborative Decision-Making Process with Simulations to Create Supply Chain Operating Plans

By working through the first challenge you have acquired some familiarity with this case and developed some mental models to help you understand how the supply chain works. Now imagine you are the person responsible for coordinating overall operations of this supply chain in a real disaster response situation. Imagine all the people wanting to talk to you, and all the information to be analyzed, and all the pressure you are under to make decisions and take action. How will you stay organized and communicate effectively with all the people involved?

As the mission director you are the person responsible for the success of this mission, yet you do not have complete authority to issue orders to all the various organizations (governmental, NGO, military, and commercial) that are participating in this humanitarian mission. So you need to use a collaborative process that focuses people’s thinking and enables quick consensus building.

[ See YouTube video on Global Logistics Cluster training program here – https://www.youtube.com/watch?v=59uLZnupB0U ]

Shown below is a concise, five-step, collaborative process can address these needs. It is both simple enough for people to understand and use under stressful conditions, and it is accurate enough to deliver the results people need to do their jobs well. It is called mission and operations planning (M&OP), and is based on the commercial supply chain practice called sales and operations planning (S&OP).

 (click on diagram for larger image)

Every mission needs a supply chain to support it, and the likelihood of success for the mission is heavily dependent on how well the supply chain meets the needs of the mission. Supply chain simulations quickly show if a particular mission is viable given the supply chain capabilities available to support it. Let’s take a quick look at how the M&OP process uses simulations to respond quickly to ever changing circumstances on a disaster response mission.

The mission forecast shows the essence of the mission and how it will operate. This information is presented in the mission’s concept of operations or “CONOPS”. The CONOPS identifies the categories of aid supplies that will be needed and where those supplies will come from. It shows where facilities will be established in the disaster area, and defines the activities that will occur at each facility. Also defined are the personnel that will be stationed at those facilities to carry out the activities. This is shown in the screenshots below.

mission-orders-and-conops (click on screenshot for larger image)

The CONOPS document provides high level overviews to communicate the key facts about the mission to a wide audience. It identifies the kinds of facilities that are needed and the location of those facilities. It also defines the functions and activities that will occur at the different facilities, and the personnel assigned to those facilities to carry out the activities. The CONOPS is a visual document that relies on maps and diagrams to convey its information (see examples in the case study introduction for the Syria Evacuation Scenario – http://blog.scmglobe.com/?page_id=4672).

Based on information in the mission CONOPS, the demand plan then specifies what products and what amounts of products are needed to support the activities and the personnel at each facility. This demand plan information is collected in the two data tables shown below.

slide3-demand-plan-1

slide4-demand-plan-2

When the demand plan is finished, it shows what products need to be delivered to what facilities during a particular period of time in order to support the activities and personnel assigned to those facilities. That provides the information needed to create the supply plan.

The supply plan shows the vehicles available to transport supplies and the routes and delivery schedules used by those vehicles as they deliver products to facilities. This supply plan information is collected in the two data tables shown below.

slide5-supply-plan-1

slide6-supply-plan-2

The demand plan and the supply plan produce the information needed to build a model of the proposed supply chain that will support the disaster response mission. The Demand plan provides the data needed for products and facilities, and the supply plan provides the data needed for vehicles and routes. The screenshot below shows how that demand and supply plan data is used to define the four entities in the supply chain model.

slide7-reconcile-plans-1(click on screenshot for larger image)

Once the supply chain model is created with data from the two plans, it can be run in simulations and the simulations will identify the discrepancies between the two plans. We can usually assume that the demand plan will remain the same, so discrepancies must be fixed by adjusting the supply plan. By doing this we are assuring that the supply chain can actually deliver the amounts of products to the facilities in time to meet the needs of the mission.

The screenshot below shows the result of one of the simulations. It shows that products are running out in Kathmandu. To respond to this and other problems that are identified by the simulations, we will adjust the supply plan. We will adjust the number of vehicles used on different routes to deliver products to different facilities.

First do whatever is needed do to get the supply chain to run for 15 – 30 days. Then go back and make adjustments to get the supply chain simulation to run for the same number of days at lower costs and lower levels of inventory.

slide8-reconcile-plans-2(click on screenshot for larger image)

After exploring different options for facilities and vehicles and routes, an effective and efficient supply chain model will emerge. The simulations provide the feedback needed to fine tune the design of the supply chain so it runs for the required number of days at a lower cost and with lower amounts of inventory.

The data for products, facilities, vehicles and routes contained in the supply chain model that produces the best simulation results defines the operating plan for the next 15 – 30 days. When the operating plan is put into action, data from daily supply chain operations is used to continually update the supply chain model. And, as shown in the screenshots below, this updated data drives new simulations to show how well the supply chain will continue to work in the coming days and weeks.

slide9-implement-and-monitor(click on screenshot for larger image)

As events unfold and situations change, use the five step M&OP to continue to monitor and modify the supply chain that supports the mission. Unexpected events occur often in disaster response situations. Being able to adapt quickly and respond effectively to unexpected events is a big part of insuring the success of the mission.

[ This five-step M&OP process is summarized in a concise 15-slide presentation that you can see and download from SlideShare here – http://www.slideshare.net/mhugos/humanitarian-mission-operations-planning-mop ]

THIRD CHALLENGE: Apply the M&OP Process to Respond to an Unexpected Situation

Test your ability to respond quickly to unexpected situations by cycling through the M&OP process to respond as situations change. Respond to these new developments described below:

  • In the midst of recovering from the first earthquake, a second earthquake hits and product demand increases by 50 percent at the Dhading and Barbare warehouses.
  • In addition, two new local warehouses with the same size and product demands as Barbare need to be set up and supplies delivered to them. The first one is in the town of Urleni north of Kathmandu, and the second is in the city of Pokhata northwest of Kathmandu.

For added realism, do this exercise online with several teams representing different organizations (governmental, NGO, military, and commercial). The teams participate online from geographically dispersed locations. And one person plays the role of the mission director who coordinates the collaboration between teams in order to quickly produce a new supply chain operating plan.

Structure and Timeline of Online Training Exercise

Teams practice working together to update mission logistics plans to respond to changes in the mission. Initial DR supply chain presented here is starting point for the exercise. Mission director moderates online exercise. Tight timeline for the exercise emphasizes need for quick decision making in disaster response situations.

The mission director prepares the updated CONOPS document prior to the start of the exercise and sends it out to all the participants. Upon receipt of the updated CONOPS document, all participants log onto an online web conference where everyone can see the mission director’s computer screen which is showing the existing supply chain as illustrated in the diagram below. The teams are composed of people who are in different geographical locations. Total time for this exercise is approximately 90 minutes.

cloud-based-collab-platform-nepal(click screenshot to see larger image)

Mission director presents updated CONOPS and answers questions from teams.
(10 – 15 minutes)

Demand team updates the demand plan — Mission director moderates this discussion and enters team’s updated demand data into the model.
(10 – 15 minutes)

Supply team specifies vehicles and routes to meet new demand — This updated supply data is also entered into the model.
(10 – 15 minutes)

Run supply simulations — Data entered from Demand team and Supply team defines the initial supply chain model to support the mission. Run simulations to find out when and where mismatches will occur between demand at facilities and supplies being delivered.
(done immediately after entry of data from supply team)

Mission director moderates group discussion — Reconcile demand plan and supply plan. Group finds how best to deliver needed supplies and keep the supply chain running for 15 – 30 days. Mission director challenges teams to lower transportation and operating costs and reduce inventory. Keep discussions focused, and enable rapid modeling and simulation of different ideas. Pick best ideas.
(20 – 30 mins)

Exercise debriefing

  • Where did problems and delays occur that slowed down the process?
  • What could be done to keep the process flowing in a timely fashion?
  • Describe what worked well and explain why it worked well.
  • Discuss how this process might improve operations and outcomes in real disaster response missions.

(15 minutes)

[This 5-step M&OP process is summarized in a concise 15-slide presentation that you can see and download from SlideShare here – http://www.slideshare.net/mhugos/humanitarian-mission-operations-planning-mop]

 

Copyright © 2016 by SCM Globe Corp.