Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Advertisement

Citing this Article

Right click to copy or hit: ctrl+c (cmd+c on mac)

Published on 25.06.20 in Vol 8, No 6 (2020): June

Preprints (earlier versions) of this paper are available at http://preprints.jmir.org/preprint/19202, first published Apr 07, 2020.

This paper is in the following e-collection/theme issue:

    Original Paper

    Medical Emergency Resource Allocation Model in Large-Scale Emergencies Based on Artificial Intelligence: Algorithm Development

    School of Information Science and Engineering, Qilu Normal University, Jinan, China

    Corresponding Author:

    Lin Du, MSc

    School of Information Science and Engineering

    Qilu Normal University

    No 33, Shanshi East Road

    Jinan

    China

    Phone: 86 13793161610

    Email: dul1028@163.com


    ABSTRACT

    Background: Before major emergencies occur, the government needs to prepare various emergency supplies in advance. To do this, it should consider the coordinated storage of different types of materials while ensuring that emergency materials are not missed or superfluous.

    Objective: This paper aims to improve the dispatch and transportation efficiency of emergency materials under a model in which the government makes full use of Internet of Things technology and artificial intelligence technology.

    Methods: The paper established a model for emergency material preparation and dispatch based on queueing theory and further established a workflow system for emergency material preparation, dispatch, and transportation based on a Petri net, resulting in a highly efficient emergency material preparation and dispatch simulation system framework.

    Results: A decision support platform was designed to integrate all the algorithms and principles proposed.

    Conclusions: The resulting framework can effectively coordinate the workflow of emergency material preparation and dispatch, helping to shorten the total time of emergency material preparation, dispatch, and transportation.

    JMIR Med Inform 2020;8(6):e19202

    doi:10.2196/19202

    KEYWORDS



    Introduction

    After an emergency, enough emergency supplies should be sent to the disaster area as soon as possible to minimize the casualties. These emergency supplies include medical equipment, medicine, food, drinking water, and protection from the cold. An important factor that affects the length of time needed to deliver emergency supplies is the amount of advance preparation. If most of the emergency supplies are already prepared in peacetime, they can be directly transported to the disaster area when the emergency arrives, saving time in accumulating supplies. The government's work in the preparation of emergency supplies generally follows a particular process. First, the government predicts possible future emergency needs based on relevant information such as the type, time, location, intensity, and probability of future emergency events, which is provided by relevant scientific research institutions. Then, it obtains funds for emergency materials through various methods. Next, it contacts relevant companies and purchases emergency materials. Finally, it stores the emergency materials that have been purchased. The government's daily preparation of emergency supplies is not without cost, and it should consider the two goals of ensuring safety and reducing expenditure.

    Therefore, the first purpose of this paper is to solve the problem of the type and quantity of emergency materials prepared in advance. The second purpose of this paper is to develop an internet-based decision support platform that can perform that function. Whether the government is judging the types and quantities of emergency materials demanded or researching emergency material scheduling strategies, it must rely on experts in related fields, including geological experts, astronomical experts, and other experts related to the nature of the emergency. It also includes medical experts and emergency management experts who need to be involved regardless of the type of emergency. Gathering these experts and government decision makers in the same place in a short time is basically impossible, and there is no need to do so to carry out decision-making work. The use of the internet can solve this problem by achieving a virtual assembly of experts and government decision makers.


    Methods

    Preparation, Dispatch, and Transportation of Emergency Materials for Major Emergencies

    Coordinated Operation of Emergency Material Preparation and Transportation

    After determining the need and demand for the type of emergency supplies, the shipping process begins. In major emergencies, the emergency supply–shipping process consists of 3 subprocesses: preparation of emergency supplies, scheduling of emergency supplies, and transportation of emergency supplies. First, when the government receives a new demand for emergency supplies, the government emergency response decision-making group needs to prepare these emergency supplies through various channels. Then, it convenes an expert group to analyze the current traffic information, material transport routes, and expected transit time, and it uses computers to calculate optimal vehicle transportation routes. Finally, vehicles carry emergency supplies to the affected area and in accordance with a predetermined route.

    The transportation of emergency supplies also has 2 subprocesses: the first is the preshipment of emergency supplies, and the second is the shipment of emergency supplies once the schedule has been finalized. After the government demands preparation of emergency supplies, the staff responsible immediately begin to prepare the supplies without having to consider the progress of the other subprocesses. As a result, the preparation subprocess in not the entire critical path.

    In the process of preparing, scheduling, and transporting medical supplies, the subprocesses of preparing emergency supplies, scheduling their delivery, and transporting them are carried out simultaneously. Because both the distance between the hub and each disaster and the number of vehicles involved in the transport are already determined, the optimal transit times of emergency supplies cannot be changed [1].

    Subprocess of Emergency Material Preparation for Major Emergencies

    According to the previous analysis, in major emergencies there are 3 main channels for the government to prepare emergency materials. First, the government directly dispatches the prepared emergency materials in stock. Second, the government supplements and purchases emergency materials from suppliers. Third, the government asks the media to request that the public donate emergency supplies.

    In the first channel, the cost of accessing the stocks of emergency materials is much less than the cost of urgently sourcing the same emergency materials from suppliers. In addition, after receiving the transfer notification regarding the stock, the government can immediately ship the emergency materials without spending time in other areas. Finally, the place of storage for the emergency supply stocks is either the distribution center or very close to the distribution center, so the transportation time is also much shorter than that of the other material preparation channels. However, if the emergency supplies in stock cannot meet the rescue needs, the government can use the 2 other channels to prepare emergency supplies. The most widely used channel is ordering emergency materials from suppliers. Requesting that the public donate emergency supplies is relatively rare. Because the time of arrival, quantity, and quality of emergency materials obtained in this way cannot be controlled by the government, they cannot be included in the unified planning.

    To sum up, these 3 channels together comprise the subprocess of emergency material preparation for major emergencies.

    Subprocess of Emergency Material Dispatch for Major Emergencies

    There are 2 key steps that need to be completed in order to dispatch emergency supplies. The first step is expert evaluation, and the second step is computer calculation [2].

    First, the experts comprehensively evaluate real-time road conditions based on aerial photographs, information provided by local governments, and feedback information from drivers of transportation vehicles, thereby determining the best route between disaster sites, as well as the required transportation time. The expert's evaluation is very important because the result of the evaluation determines the optimal transportation strategy. However, in order to ensure the continuity of the entire process, it is necessary to strictly limit the time used by expert evaluation. For a cluster decision, which is made by multiple experts independently, experts generally make one or more rounds of judgment before making a final decision. Under this scenario, it obviously saves more time if experts make only 1 round of judgment. Since each expert judges the existing information differently, it is possible for them to evaluate each possibility of road conditions. In order to make the conclusion drawn by the experts conform to a unified standard, the increase in the expected transportation time between 2 affected areas compared with original transportation time between them can be set to m possible minutes. Then experts can evaluate the probability of these transportation times. If there are n experts participating in the evaluation, the probability of the i expert evaluating the m cases is Pi(L1),Pi(L2),…Pi(Lm). According to the evidence synthesis algorithm, the expert's evaluation of the probability of the j case is:

    See Equation 2:

    As a result, the latest expected increase in time for the vehicle to travel a certain route is the situation with the highest evaluation probability. Immediately afterwards, the latest traffic information is substituted into the material transportation scheduling model proposed above and the optimal transportation strategy can be calculated. This calculation process is completed by computer [3].

    Models for Emergency Material Preparation, Dispatch, and Transportation

    Workflow System for Emergency Material Preparation, Dispatch, and Transportation

    If the time for vehicle k + 1 and vehicle k to return to the distribution center is t(k+ 1) and tk, respectively, and vehicle k + 1 returns after vehicle k, and if there is no other vehicle between the 2 vehicles, then the time interval is t(k+ 1) - tk. tp refers to the transportation of materials. If the total number of vehicles participating in transportation is m, the minimum time interval for all returning vehicles is:

    tp = min [t2t1,…t(k+ 1) - tk,…tm - t(m- 1)] (3)

    The subprocesses for emergency supply preparation and emergency supply scheduling are together called the material preparation and scheduling process. Each batch of materials must go through this material preparation and scheduling process in order to be transported to the disaster area. If there is a failure, the government must look for other suppliers to enter the next round of negotiations. This process refers to the time period required for this meeting [4].

    Here, tTN refers to the required time, and E refers to the set of optimal solutions. This formula is to calculate the optimal solution in the shortest time:

    The emergency materials prestored in the government warehouse can be shipped immediately after getting the transportation instructions. The preparation time is 0. The time-consuming part of the material preparation subprocess is the following: the government directly dispatches the stock emergency materials prepared in advance, the government supplements and purchases emergency materials from suppliers (if necessary), and the government requests the society to donate emergency materials through the media (if necessary). The government waits for the maximum value of the three processes. The total running time of this material preparation and scheduling process is shown below:

    In this equation, tTN is the time used by the government to negotiate with the supplier, tTP is the time spent by the supplier to prepare materials after the negotiation is successful, tTM is the time required for the media to publish a report, and tTD is the time between the public seeing the media report and preparing to donate. Regarding the time to receive materials, tTL is the average time for emergency materials to be transported from the supplier to the distribution center, tTE is the expert's estimated transit time of each transportation route, and tTA is the computer calculation using the emergency material transportation scheduling model.

    According to the previous analysis, this material preparation and dispatch process should be completed within the minimum time interval between the 2 transport vehicles arriving at the material distribution center:

    tp = t'p (6)

    Set in the period of [0,T1], where T1 equals the first time, the time of occurrence of each demand for emergency supplies is independent and follows a negative exponential distribution. During this period, when urgent materials are needed, we can know the average arrival rate of demands for emergency supplies:

    The average cost of demands for emergency supplies is

    According to the principle of queueing theory, the length of stay follows a negative exponential distribution with parameter μ−λ. Then, the probability density of the dwell time w is:

    The average length of stay can be determined as follows, where w is the final result:

    The average waiting time can be found below:

    In addition, emergency materials, such as medical supplies and living necessities, must be delivered within a certain period of time. Otherwise, it will cause greater casualties. To prevent this, the formula can be set so that these emergency materials must be delivered within the [0,T1] period. Among the formulas, suppose that a certain link i in the material preparation and scheduling process needs n staff to participate together, and the amount of time of this link is tpi. The total effective working time of the n staff members is ntpi. cpi refers to a single specific index and Rp refers to the composite index. If the time utilization rate of these staff is set to rp(0<rp<1), then the number of staff required is

    Using the Whooped (Microsoft Corp) software to conduct a static test on the system, the system has passed the coverage, boundedness, and activity tests, so the design of this system is reasonable [5].

    Simulation System Framework for Emergency Material Preparation and Dispatch

    Based on the above ideas, this paper establishes a simulation system framework for emergency material preparation and dispatch, as shown in Figure 1. The simulation system framework presents all links in the material preparation and material transportation subprocesses in the first-level system, ignoring the difference in the sourcing and preparation time of each batch of emergency materials and removing all emergency materials in the government emergency process [6].

    Figure 1. Framework of emergency material preparation and dispatch simulation system. PC: PC-BSD installer; PD: product department; PDR: project definition report find suppliers; PE: portable executable; PF: planned finish date; PL: pay local; PN: part number; PNR: prior notice required; PR: pattern recognition; PS: performance standard; PW: present worth; PWR: process work request; TA: computer computes optimal transportation routes; TC: trace cache; TD: corporate donations of emergency supplies; TE: experts assess the trip transport routes by the expected time; TI: the government gives the current traffic information; TL: set off emergency supplies destined for distribution center; TM: media reports; TN: government negotiations with suppliers; TP: suppliers prepare emergency supplies; TS: find suppliers; TW: government warehouse to prepare emergency supplies.
    View this figure

    Results

    Radio frequency identification technology can be mainly applied to the storage of emergency supplies. It is connected to a database, and its functions include the storage and inventory management of emergency materials, as well as the management of emergency materials after major emergencies. In daily situations, the government puts the emergency materials into storage after purchasing them. During this time, the radio frequency identification technology can automatically identify and store the materials according to their category and storage environment requirements. Additionally, the long-term storage of materials is generally limited by shelf life. This technology can automatically pick out materials that have passed their shelf life and prompt management personnel to replace them with new materials. Moreover, as the amount of emergency materials stored changes, the material storage warehouse may also need to adjust its structure or expand its scale. At this time, the technology can help managers find the materials that need to be transferred and ensure the orderly progress of the entire transfer [7].

    After major emergencies occur, emergency supplies need to be transported out of the warehouse to the affected area. Radio frequency identification technology can enable managers to track the status of each emergency item at any time and help them formulate the best transportation plan. More importantly, the use of this technology allows managers to more quickly calculate the difference between the inventory and the demand for various emergency supplies.

    The application of the Internet of Things technology in the process of emergency material preparation and transportation in major emergencies is mainly achieved through multiple terminal devices of different types, such as active and passive microwave transmissions, reception devices involved in radio frequency identification technology, GPS terminal signal receivers, ground control instruments, and satellites. The most important role played by these terminal devices is to facilitate the collection of multiple types of decision data. However, in the government emergency process developed in this paper, a large amount of data entry is required. Therefore, the combination of the decision support platform with the Internet of Things technology is feasible and even efficient.

    Based on this idea, this paper proposes a framework for the organic integration of the related technologies of the Internet of Things and the government emergency process [8]. After the integration, the new decision support platform will effectively integrate emergency supplies, rescuers, decision makers, and computer artificial intelligence into a large system, which will significantly increase the speed of information transmission and processing, thereby greatly improving the operational efficiency of preparation, scheduling, and transportation of emergency supplies [9].


    Discussion

    This paper establishes a model for emergency material preparation and dispatch based on queueing theory; further establishes a workflow system for emergency material preparation, dispatch, and transportation based on a Petri net; and provides a simplified simulation system framework for emergency material preparation and dispatch with high operating efficiency. It can effectively coordinate the workflow of emergency material preparation and dispatch, shortening the total time needed by these processes. Based on the Internet of Things technology that can be used in the process of emergency material transportation, an integrated framework for emergency material financing is constructed. This paper also provides an interface to the model solution software on the decision support platform so that this decision support platform integrates all the principles and algorithms of emergency material financing and transportation proposed in this paper.

    Acknowledgments

    The author acknowledges the Key Research and Development Program of Shandong Province (Grant No. 2019GSF108010).

    Conflicts of Interest

    None declared.

    References

    1. Tucci VT, Moukaddam N, Alam A, Rachal J. Emergency Department Medical Clearance of Patients with Psychiatric or Behavioral Emergencies, Part 1. Psychiatr Clin North Am 2017 Sep;40(3):411-423. [CrossRef] [Medline]
    2. Eaton-Evans T. Managing medical emergencies at endurance rides. In Practice 2019 Jul 11;41(6):270-274. [CrossRef]
    3. Zhou QS, Olsen TL. Inventory rotation of medical supplies for emergency response. European Journal of Operational Research 2017 Mar;257(3):810-821. [CrossRef]
    4. Ki M. Surveillance and epidemiologic investigation in public health emergencies caused by infectious diseases. J Korean Med Assoc 2017;60(4):292. [CrossRef]
    5. Schönfeldt-Lecuona C, Gahr M, Schütz S, Lang D, Pajonk FGB, Connemann BJ, et al. Psychiatric Emergencies in the Preclinical Emergency Medicine Service in Ulm, Germany in 2000 and 2010, and Practical Consequences. Fortschr Neurol Psychiatr 2017 Jul;85(7):400-409. [CrossRef] [Medline]
    6. Nable JV, Tupe CL, Gehle BD, Brady WJ. Is there a doctor on board? In-flight medical emergencies. Cleve Clin J Med 2017 Jun;84(6):457-462 [FREE Full text] [CrossRef] [Medline]
    7. Albelaihi H, Alweneen A, Ettish A, Alshahrani F. Knowledge, Attitude, and Perceived Confidence in the Management of Medical Emergencies in the Dental Office: A Survey among the Dental Students and Interns. J Int Soc Prev Community Dent 2017;7(6):364-369 [FREE Full text] [CrossRef] [Medline]
    8. Aftergood DE. In-Flight Medical Emergencies. N Engl J Med 2016 Jan 21;374(3):292. [CrossRef] [Medline]
    9. Goertzel B. Artificial general intelligence. Lecture Notes in Computer Science 2017;56(2):32-39. [CrossRef]

    Edited by H Li, Z Du, K Kalemaki; submitted 07.04.20; peer-reviewed by Z Liu, Y Liu, M Zhang; comments to author 15.04.20; revised version received 22.04.20; accepted 23.04.20; published 25.06.20

    ©Lin Du. Originally published in JMIR Medical Informatics (http://medinform.jmir.org), 25.06.2020.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Medical Informatics, is properly cited. The complete bibliographic information, a link to the original publication on http://medinform.jmir.org/, as well as this copyright and license information must be included.