Hewlett-Packard: Using Network Visualization to Mitigate Supply Chain Risks
I recently spoke to Trace White and Travis Parker from HP Inc.(HP) about using supply chain network design and visualization tools to respond to supply chain risks. This is something that their team at HP started to examine about two to three years ago. They had built a supply chain visualization tool in response to getting a better handle on how to position supply chain initiatives, such as direct ship activities. At the time, HP had visibility into their supply chain, but they lacked the speed to connect an event to the network and respond appropriately. Basically, they could see an event was happening, but were unable to make changes to their network to mitigate the damage. The company knew what to do but could not get the information quick enough.
While HP was trying to address speed, they also saw a rise in disruptions. They realized that climate–related issues have gone up significantly in recent years. Considering that the company was running lean, these disruptions made the risks even worse. So the team took the visualization database they had built and added a risk management layer. To date, the tool has been up and providing real practical experience for nine months.
So how does the process work? HP feeds threats from third party data sources into the tool. Analysts monitor and evaluate whether threat will impact the supply chain. The biggest challenge for HP is the data – they have a pretty complete set of data of manufacturers and suppliers, but it’s not perfect. The company is very confident with the data they have for Tier 1 suppliers, but the further down you go, the accuracy gets more challenging. This data is essential for contacting HP manufacturing sites as well as suppliers if there is going to be a disruption for products they sell directly to consumers.
According to HP, the company uses two main subscription sources for data threats: NC4 for supply chain disruptions and Anvil for travel and security. The company also uses assorted services to bring in financial or business issues, especially in Asian geographies. The main types of disruptions the company monitors are natural disasters, legal and financial disruptions, infrastructure, health, and social and political unrest. For example, if there is an earthquake, HP can map the threat and see what the basic radius of impact is. They can draw the radius onto the map and view a list of all sites (HP manufacturing and suppliers ((down to tier 2-3))) that could be impacted. At this point, the company can call or email a survey to all site contacts for a status update. Before, it would take 12-24 hours to pull the list and make contact with sites. Now, it takes as little as a few hours. Here again is where accurate and up-to-date data on each site is critical.
HP uses many analytics tools to monitor and react to these supply chain disruptions. For example, the company uses LLamasoft data for airports and seaports. This data helps the company make supply chain decisions if there are port strikes or airport closures. HP also uses Supply Chain Guru for network design and optimization, and risk management. Using risk assessments and probabilities gives HP a smarter look at designing supply chains. The company is also tagging analytics onto it for better analysis.
The visualization tool is enabling HP to run smarter, leaner, and more efficient. One finding from all the risk assessment and analysis was that not all locations are as valuable, and that is something that has to be considered. A prime example of this is when there are disruptions at a shipping facility and merchandise needs to be shipped from another location. The location that is the closest and can provide a shorter delivery route may not always be the best resource for fulfilling an order. This is especially true if the order will now have to pass through a dangerous or high-theft location. At this point, it makes more sense to minimize this risk and opt for a longer shipping route.
A recent example of HP putting the tool to use was during the flooding in Chennai India in 2015. There were all kinds of shipping delays due to the floods. The bigger problem for HP was flooding to their customer service and call center. Using the tool, HP was able to identify groups that were impacted, move these groups of employees to hotels, and run operations from there. The logistics team started establishing alternate shipping lanes to get orders out as soon as possible. Within 18 hours of assessment, the company had re-routed the pipeline of HP products and services to safer routes and re-located flooded HP global call center employees to hotel to service customers. While there were long hold times up to 5 hours at first, HP resolved the problem in a much faster timeframe than normal.
In conclusion, HP has designed a tool to respond to the rise of global supply chain disruptions. The solution uses a centralized supply chain node database with graphing and visualization tools. This gives the company real-time visibility into supply chain disruptions. As a result, HP is able to respond to disruptions in a timely manner, usually within 12 to 24 hours. The result is that global operations can remain up and running and customers are still getting high service levels. This enables HP to mitigate the cost of future disruptions. The next phase of the solution will be to enable the use of predictive analytics and add cyber security as well.
Visit website to view article
Aggarwal, P. and K. Moinzedeh. (1994) Order Expedition in Multi-Echelon Production/Distribution Systems. IIE Transactions, Vol. 26, 86–96.CrossRefGoogle Scholar
Allen, S. and D. Esopo. (1968) An Ordering Policy for Stock When Delivery Can Be Expedited. Operations Research, Vol. 16, 830–833.CrossRefGoogle Scholar
Axsater, S. (1993) Exact and Approximate Evaluation of Two-Level Inventory Systems. Operations Research, Vol. 41, 777–785.Google Scholar
Bulinskaya, E. (1964) Some Results Concerning Optimal Inventory Policies. Theory of Probability and Applications, Vol. 9, 389403.Google Scholar
Bulinskaya, E. (1964) Steady-State Solutions in Problems of Optimal Inventory Control. Theory of Probability and Applications, Vol. 9, 502–507.CrossRefGoogle Scholar
Chiang, C. and G. Gutierrez. (1996) Periodic Review Inventory System with Two Supply Modes. European Journal of Operational Research 94, 527–547.CrossRefGoogle Scholar
Clark, A.J. and H. Scarf. (1960) Optimal Policies for a Multi-Echelon Inventory Problem. Management Science, Vol. 6, 475–490.Google Scholar
Daniel, K. (1962) A Delivery-Lag Inventory Model with Emergency Order. In: Multistage Inventory Models and Techniques,Scarf, Gilford, Shelly (Eds.), Stanford University Press, Stanford, California.Google Scholar
Deuermeyer, B. and L. Schwarz. (1981) A Model for the Analysis of System Service Level in Warehouse Retailer Distribution Systems. In: Multi-Level Production/Inventory Control Systems: Theory and PracticeSchwartz, L. (Ed.), TIMS Studies in Management Science 16Amsterdam, North-Holland 51–67.Google Scholar
Federgruen, A. and P. Zipkin. (1984) Allocation Policies and Cost Approximations for Multilocation Inventory Systems. Naval Research Logistic Quarterly, Vol. 31, 97–129.Google Scholar
Federgruen, A. and P. Zipkin. (1984) Approximations of Dynamic Multilocation Production and Inventory Problems. Management Science, Vol. 30, 69–84.Google Scholar
Fukuda, Y. (1960) Optimal Policies for the Inventory Problem with Negotiable Leadtime. Management Science, Vol. 10, 690–708.Google Scholar
Graves, S. (1985) A Multi-Echelon Inventory Model for a Repairable Item with One-For-One Replenshment. Management Science, Vol. 31, 1247–1256.Google Scholar
Graves, S. (1996) Multiechelon Inventory Model with Fixed Replenishment Intervals. Management Science, Vol. 42, 1–18.Google Scholar
Jackson, P. (1988) Stock Allocation in a Two-Echelon Distribution System, or `What To Do Until Your Ship Comes In’. Management Science, Vol. 34, 880–895.Google Scholar
Karlin, S. (1960) Dynamic Inventory Policy with Varying Stochastic Demands. Management Science, Vol. 6, 231–258.Google Scholar
Lee, H. and K. Moinzadeh. (1987) Two Parameter Approximations for Multi-Echelon Repairable Inventory Models with Batch Ordering Policy. IIE Transactions, Vol. 19, 140–149.Google Scholar
Moinzedeh, K. and H. Lee. (1986) Batch Size and Stocking Levels in Multi-Echelon Repairable Systems. Management Science, Vol. 32, 1567–1581.CrossRefGoogle Scholar
Moinzedeh, K. and S. Nahmias. (1988) A Continuous Review Mode for an Inventory System with Two Supply Modes. Management Science, Vol. 34, 761–773.Google Scholar
Moinzadeh, K. and C. Schmidt. (1991) An (S — 1, S) Inventory System with Emergency Orders. Operations Research, Vol. 39, 308–321.Google Scholar
Neuts, M. (1964) An Inventory Model with an Optional Time Lag. SIAM, Vol. 12, 179–185.Google Scholar
Pyke, D. and M. Cohen. (1994) Multiproduct Integrated Production-Distribution Systems. European Journal of Operations Research, Vol. 74, 18–49.Google Scholar
Sherbrooke, C. (1968) Metric: Multi-Echelon Technique for Recoverable Item Control. Operations Research, Vol. 16, 122–141.Google Scholar
Svoronos, A. and P. Zipkin. (1988) Estimating the Performance of Multi-Level Inventory Systems. Operations Research, Vol. 36, 57–72.Google Scholar
Veinott, A. (1965) Optimal Policy for a Multi-Product, Dynamic, Nonstationary Inventory Problem. Management Science, Vol. 12, 206–222.Google Scholar
Veinott, A. (1966) The Status of Mathematical Inventory. Management Science, Vol. 12, 745–777.Google Scholar
Whittmore, A. and S. Saunders. (1977) Optimal Inventory Under Stochastic Demand with Two Supply Options. SIAM Journal of Applied Mathematics, Vol. 32, 293–305.Google Scholar
Wright, G. (1968) Optimal Policies for a Multi-Product Inventory System with Negotiable Leadtimes. Naval Research Logistics Quarterly, Vol. 15, 375–401.Google Scholar