Computational Analytics of Value Networks - Assignment Example

Computational Analytics for Value Networks (Individual) Student Name: Course: 3rd June, 2017 DEA analysis describes a performance measurement technique used to evaluate the performance of different decision-making units (DMU). It helps decision makers to: • Calculate an efficiency score for a firm. This helps determine whether a firm is efficient or has room to improve. • Determines how input needs to be added or how much output needs to be reduced in order to make a firm efficient. • Identify the nature of returns of a firm to its scale in order to determine how the scale needs to change for the firm to reduce average cost. • Identify a set of benchmarks, which can be used to improve the rest of the firm’s processes by comparing against the benchmarks. In our case, we are analyzing the productivity of 20 hospitals using the data provided. A DEA analysis was conducted for the 20 hospitals using the DEAP analysis software. Each of the different scenarios is discussed below with the implications of each result used to advice on the managerial and operational areas for improvement as appropriate. The following acronyms will be used for interpreting the data. • VRS – Variable Returns to Scale • TE – Technical Efficiency • SE – Scale Efficiency Case 1: Use DEA where goal is to minimize inputs In the case of an output-oriented model, it is assumed that the hospital manager has more control on the inputs rather than the number of patients arriving at the hospital for acute or minor cases. Table 1 shows the calculation of benchmarking for peers of each hospital using input-oriented VRS efficiency for all twenty NSW Health hospitals. Table 1: Results of DEA using DEAP for input – oriented VRS Model   -- Efficiency -- --Benchmarks-- DMU No. (Hospital) Input-Oriented VRS Efficiency Peer 1 Weight (λ) Peer 2 Weight (λ) Peer 3 Weight (λ) Peer 4 Weight (λ) 1 0.885 12 0.103 15 0.897         2 0.360 12 0.167 15 0.833         3 0.770 15 0.810 6 0.190         4 0.626 15 0.378 6 0.622         5 1.000 6 1.000             6 1.000 6 1.000             7 0.561 12 0.049 6 0.319 15 0.632     8 1.000 8 1.000             9 1.000 9 1.000             10 0.934 11 0.818 6 0.182         11 1.000 11 1.000             12 1.000 12 1.000             13 0.976 12 0.500 11 0.500         14 0.718 12 0.048 15 0.779 6 0.173     15 1.000 15 1.000             16 1.000 16 1.000             17 1.000 17 1.000             18 0.990 12 0.069 15 0.729 6 0.202     19 0.737 16 0.103 15 0.268 8 0.471 6 0.158 20 0.926 9 0.544 17 0.075 15 0.381     Each lambda weight associated with the peers of each hospital points to its relative importance in the peer group, at least according to an explanation offered by Huguenin (2012). In the case of Hospital 1, best practice analysis should be formed by a composite hospital formed by 10.3% of Hospital 12, and 89.7% of Hospital 15. The ‘virtual’ hospital does not exist and thus Hospital 1 should concentrate its benchmarking practice analysis on the peer associated with the highest lambda weigh, in this case Hospital 15. This should apply to all hospitals in the case of minimizing inputs. Hospitals 5, 6, 8, 9, 11, 12, 15, 16, and 17 are found to be technically efficient. The categories of technical efficiency in this case are the productivity index of medical personnel at the hospital especially doctors and nurses, the utilization of beds in the facility, the efficiency of its dispatch department, and the efficiency of emergency services. These hospitals act as peers to other hospitals in how input can be minimized to achieve technical efficiency. Table 2 shows both input and output slacks or improvements that can be done for the different hospitals. Table 2: Slacks for Input-oriented VRS Model   -- Input Slacks -- --Output Slacks-- DMU No. (Hospital) Nurses Beds Minor Cases Acute Cases 1 0.000 210.385 155.128 0.000 2 0.000 66.667 83.333 0.000 3 0.000 0.000 167.209 12.362 4 133.000 0.000 0.000 10.444 5 180.000 0.000 25.000 25.000 6 0.000 0.000 0.000 0.000 7 0.000 0.000 30.799 0.000 8 0.000 0.000 0.000 0.000 9 0.000 0.000 0.000 0.000 10 86.860 0.000 50.000 0.000 11 0.000 0.000 0.000 0.000 12 0.000 0.000 0.000 0.000 13 93.824 0.000 75.000 0.000 14 0.000 0.000 43.395 0.000 15 0.000 0.000 0.000 0.000 16 0.000 0.000 0.000 0.000 17 0.000 0.000 0.000 0.000 18 0.000 0.000 27.989 0.000 19 0.000 0.000 0.000 0.000 20 0.000 122.702 0.000 0.000 Mean 24.684 19.988 32.893 2.390 To become technically efficient, Hospital 1 needs to reduce its bed capacity by 210.385 beds. It can also achieve technical efficiency by improving its operations with the same input to ensure it addresses an additional 155.128 minor cases. The same goes for Hospital 2 that needs to reduce its bed capacity by 66.667 beds or improve the productivity of inputs and cater for an additional 83.333 minor cases. The same can be interpreted for the rest of the hospitals with each input slack value representing a needed decrease in input while each output slack value represents a needed increase in outputs for technical efficiency to be achieved. Table 3 shows the efficiency input and output targets for an input-oriented VRS model Table 3: Efficiency input and output targets for input-oriented VRS model Efficiency Input target Efficiency Output target DMU (Hospital) Nurses Beds   Minor cases Acute cases 1 177.051 320.769   305.128 50.000 2 215.833 365.000   308.333 75.000 3 153.988 230.982   257.209 22.362 4 242.556 187.778   160.000 50.444 5 320.000 150.000   75.000 75.000 6 320.000 150.000   75.000 75.000 7 210.234 252.281   230.799 50.000 8 400.000 320.000   350.000 100.000 9 550.000 480.000   400.000 90.000 10 753.636 616.364   300.000 300.000 11 850.000 720.000   350.000 350.000 12 720.000 940.000   350.000 400.000 13 785.000 830.000   350.000 375.000 14 179.597 265.804   263.395 40.000 15 115.000 250.000   300.000 10.000 16 600.000 590.000   320.000 275.000 17 550.000 710.000   375.000 230.000 18 198.096 277.334   257.989 50.000 19 331.682 302.199   290.000 90.000 20 384.156 409.563   360.000 70.000 The figures in Table 3 show the projected inputs and outputs for an input-oriented VRS if the hospitals are to be considered technically efficient. Case 2: Use DEA where goal is to maximize outputs Here, it is assumed that hospital managers through various initiatives like marketing, referrals, improvement in the efficiency of execution of emergency services, dispatch services and other means can attract both minor and acute case patients to the hospital. It is also assumed they can influence the efficiency of admitting and discharging these patient as shown by Ozcan (2008). Table 4 shows the technical efficiency of each hospital and benchmarking for peers of each hospital using output-oriented VRS efficiency for all twenty NSW Health hospitals. Again, lambda weights associated with peers of each hospital points to their relative importance in benchmarking for technical efficiency. Table 4: Results of DEA using DEAP for output-oriented VRS Model   Efficiency --Benchmarks-- DMU No. (Hospital) Input-Oriented VRS Efficiency Peer 1 Weight (λ) Peer 2 Weight (λ) Peer 3 Weight (λ) Peer 4 Weight (λ) 1 0.772 15 0.860 12 0.140         2 0.572 12 0.133 9 0.867         3 0.284 17 0.048 9 0.098 15 0.779 8 0.075 4 0.486 15 0.185 8 0.774 6 0.042     5 0.667 6 1.000             6 1.000 6 1.000             7 0.565 9 0.378 17 0.220 15 0.402     8 1.000 8 1.000             9 1.000 9 1.000             10 0.934 11 0.895 6 0.105         11 1.000 11 1.000             12 1.000 12 1.000             13 0.988 11 0.409 12 0.591         14 0.674 9 0.135 17 0.175 15 0.69     15 1.000 15 1.000             16 1.000 16 1.000             17 1.000 17 1.000             18 0.974 12 0.073 15 0.727 6 0.201     19 0.796 17 0.135 15 0.024 9 0.243 8 0.598 20 0.979 9 0.644 17 0.045 15 0.31     Using the output-oriented VRS model, hospitals 5, 6, 8, 9, 11, 12, 15, 16 and 17 are found to be technically effect in utilizing provided inputs to produce efficient outputs. For hospitals that are not technical efficient, the inefficiency can be addresses by addressing the operation capacity of their inputs and increasing the productivity of their inputs in order to produce a higher output. In this case, vigorous marketing, positive referrals, improvement in the efficiency of execution of emergency services, efficient dispatch services and other factors can boost the number of patients being treated for both minor and acute cases at the hospital. An example is the case of Hospital 1 whose output can be boosted by 22.8% if the stated factors are addressed. This is true for each corresponding value of output-oriented VRS efficiency per hospital. Table 5 shows both input and output slacks or improvements that can be made on an output-oriented VRS model with regards to inputs and outputs for the different hospitals. From the slacks, it is seen that for Hospital 1 to be considered technically efficient, it needs to boost its bed capacity by 253.058 beds or boost its productivity to address an additional 155.128 minor cases. This interpretation can be applied for each hospital that contains slack values in Table 5. Overall, the mean values indicate that all hospitals can improve technical efficiency by adding 38.802 nurses across board and adding 54.413 beds across all hospitals. This overall technical efficiency can also be improved by boosting productivity of inputs to ensure the number of minor cases goes up by 32.893 patients while that of acute cases goes up by 2.39 cases. Table 5: Slacks for Output-oriented VRS Model   -- Input Slacks -- --Output Slacks-- DMU No. (Hospital) Nurses Beds Minor Cases Acute Cases 1 0.000 253.058 155.128 0.000 2 27.449 658.980 83.333 0.000 3 0.000 0.000 167.209 12.362 4 255.982 0.000 0.000 10.444 5 180.000 0.000 25.000 25.000 6 0.000 0.000 0.000 0.000 7 0.000 12.005 30.799 0.000 8 0.000 0.000 0.000 0.000 9 0.000 0.000 0.000 0.000 10 105.789 0.000 50.000 0.000 11 0.000 0.000 0.000 0.000 12 0.000 0.000 0.000 0.000 13 126.818 0.000 75.000 0.000 14 0.000 8.267 43.395 0.000 15 0.000 0.000 0.000 0.000 16 0.000 0.000 0.000 0.000 17 0.000 0.000 0.000 0.000 18 0.000 0.000 27.989 0.000 19 0.000 0.000 0.000 0.000 20 0.000 155.947 0.000 0.000 Mean 34.802 54.413 32.893 2.390 Table 6 shows the efficiency input and output targets for an output-oriented VRS model. For each hospital to be considered technically efficient, the following input targets or output targets need to be applied though improvement in each of the identified operational areas. Table 6: Efficiency input and output targets for output-oriented VRS model Efficiency Input target Efficiency Output target DMU (Hospital) Nurses Beds   Minor cases Acute cases 1 200.000 346.942   307.025 64.793 2 572.551 541.020   393.367 131.122 3 200.000 300.000   317.161 35.240 4 344.018 300.000   329.330 82.333 5 320.000 150.000   75.000 75.000 6 320.000 150.000   75.000 75.000 7 375.000 437.995   354.278 88.570 8 400.000 320.000   350.000 100.000 9 550.000 480.000   400.000 90.000 10 794.211 660.000   321.053 321.053 11 850.000 720.000   350.000 350.000 12 720.000 940.000   350.000 400.000 13 773.182 850.000   350.000 379.545 14 250.000 361.733   326.648 59.391 15 115.000 250.000   300.000 10.000 16 600.000 590.000   320.000 275.000 17 550.000 710.000   375.000 230.000 18 200.000 280.000   258.505 51.327 19 450.000 410.000   364.338 113.070 20 415.000 419.053   367.832 71.523 From the analysis of the above tables, it is seen that the hospitals are suffering from a low productivity index. This could be caused by a number of factors that include inputs in terms of staffing and cost, systems in use, and the hospital’s own strategies and policies. From the analysis of the 20 hospitals done, it can be concluded that the productivity index is low since they all don’t have standardized emergency equipment, strategies and policies in dealing with the minor and acute patient cases encountered in their operations. Service Value Network for New South Wales Emergency Services Service value networks (SVNs) provide crucial pathways to establishing and retaining future strong positioning within a service industry sector at least according to Hamilton (2007). The webs of relationships contained in it generate economic value through complex dynamic exchanges between organizations, individuals or groups. SVNs offer an integrated approach for investigating both online and offline services. From the scenario given, it is seen that the different emergency cases reported carry the same level of criticality. This is important since the emergency department does not wish to accommodate any delays in the provision of emergency services to the various reported cases. The main challenge the emergency department faces in the given scenario is the high number of emergency cases reported to it that seem to require the same level of attention in addressing them. To successfully handle the high number of critical cases, the emergency department needs to be equipped with adequate resources. These resources are the nurses, doctors, systems and equipment. The emergency services also include ambulances, paramedics and first responders who are dispatched to provide retrieval emergency services for patients needing medical emergency attention. To improve the operation of the service value network, inputs and outputs have to be optimized to achieve the desired results. In terms of inputs, the productivity of doctors, nurses, paramedics and first responders plays a crucial part in the efficiency of emergency services. Blaus et al (2009) state that the goal of the SVN in emergency services is to create and promote customer value and integrate accounting in order to ensure survival for the hospital, its staff, and the technology in use. Figure 1 shows the SVN mapping of emergency department Figure 1: SVN mapping for emergency services department Emergency Department Inputs and Outputs The input for the emergency department are beds, the skilled medical worker like doctors, nurses, paramedics, and clerks. Inputs also include hospital facilities like the triage, labs, treatment cubicles, recovery, and diagnostic rooms. The emergency department also has emergency dispatch services like ambulances, medical equipment, patient transportation resources and other hospital transfer services. The outputs have been optimal resources, optimal shifts, staffing resource, queue reduction strategy, reduced service time, smooth patient treatment process, adequate and effective communication, increased safety and adequate discharge process. The output process and concept has been determined through a surge of capacity and patient relationship within the patient flow that enhance implications for input and outputs. Emergency department service follows different models, but the purpose of any model is to enhance patient flow that enhances emergency department efficiency. Enhancing outputs and inputs in the emergency department require several other techniques, which involve a guide to service improvement, measurement, analysis and technique solutions. The process flow to the ED should be friendly to stakeholders who are attending to the patient from the emergence scene. Simulation modeling are also valuable resources that have been used to enhance learning, training, and information to ED services providers. High-quality skills are a resource for demanding high-quality PI. The ED should also have a system that adequately solves challenges as intimated by Jarousse (2011). The system of addressing problems should be contained in the communication flow chart for solving problems. The flow chart should indicate contact personnel and the appropriate response that can be provided. All stakeholders in the ED should validate the communication input and out process for the purpose of enhancing improved communication and improved the PI system. References Read More