Team # 41139 Page 4 of 14
α Collaborative distance between active employees and former employees Tab 1: Symbols used in this paper
2.2 Hypothesis
We make the following assumptions while formulating the manpower planning
problem to determine optimal recruitment policies. a)Overstaffing is not allowed.
b)Recruitment at a particular grade is considered.
c)Most of workers in ICM are promoted every three years.
2.3 Explanations
①Network model：The network model is a database model conceived as a flexible way of representing objects and their relationships. ②Human resource management：Human resource management is a function in organizations designed to maximize employee performance in service of their employer’s strategic objectives. ③Bayesian network:A Bayesian network is a probabilistic graphical model (a type of statistical model) that represents a set of random variables and their conditional dependencies via a directed acyclic graph (DAG). ④A Markov chain is a mathematical system that undergoes transitions from one state to another on a state space. It is a random process usually characterized as memoryless: the next state depends only on the current state and not on the sequence of events that preceded it. ⑤Churn rate：Churn rate can describe the number of employees that move within a certain period. ⑥Recruitment Cost:This is the cost incurred in the process of recruitment.The recruitment costs can be broadly classified into two categories viz..fixed costs and variable costs of recruitment which is proportional to the number of people recruited.The following is the different components of recruitment cost. A)Cost of Advertising
B)Cost of administrative authority which makes recruitment policy C)Cost of manpower working on the processing of applications D)Cost of information processing E)Cost of conducting written test
Team # 41139 Page 5 of 14
F)Costs incurred in the form of payment to the interview committee members(if hired
from outside) or the wages of the people in the interview committee(if the members are internal)
G)Traveling expenses paid to the candidates
H)Cost of medical examination done by the organization ⑦Understaffing Costs:Understaffing costs are the costs resulting due to decreased productivity and loss of goodwill(in a profit motivate organization) as a result of the non-availability of the workforce. ⑧Training Cost:This is the cost for training.
3. Models
3.1 The human capital network model(we use this model into task#1)
Assumption 1: ①Promotion within an office: Everyone has the possibility of promotion. But the promotion is only allowed in the different departments, the experienced employee were also not allowed to have the ability with the Ministry of promotion ②Promotion between the offices: Mobilize the same level of office is not considered for promotion ③Promotion extensive :Experienced staff can promoted to no experience supervisor, experienced executives can promoted to junior managers. Now, build the HR network model in ICM organization.
Team # 41139 Page 6 of 14
Fig1: Staff cost allocation system
Explanation:
A=Senior manager B=Junior manager
C=Experienced supervisor D=Inexperienced supervisor E=Experienced employee F=Inexperienced employee G=Administrative clerk
Assumption2:
Based on past experience analysis, the topology relationship of the network is given using the 370*370 adjacent matrix below:
?a11a12?aa2221?A????an1an2a1n? a2n????ann??0There is no promotion from i to jWhere aij??
?1There is promotion from i to j
Consider a variety of factors, such as the turnover rate or promotion system, we designed this manpower system model.In other words,we build the HR model involving both promotion and new entry.
3.2 The Bayesian network Model(we use this model into task#2)
Team # 41139 Page 7 of 14
Fig2:Bayesian network
A Bayesian network is a probabilistic graphical model (a type of statistical model) that represents a set of random variables and their conditional dependencies via a directed acyclic graph (DAG).We use it express the relationship among the factors in Fig2.And the formula is as follows:
P(Y|x1,x2,...,x8)?P(Y)P(x1,x2,...x8|Y)?P(Y)P(x,x,...,xj12j?1k8|Yj) where
P(x1,x2,...,x8|Y)?P(x1|x5)P(x2)P(x3)P(x4|x5)P(x5)P(x6)P(x7)P(x8)
We assumed the churn rate of various factors as follows from Tab2 —Tab8:
Resign Or Not Yes No Years of work <3 3~6 6~9 9~12 12~15 >15 10% 90% 20% 80% 40% 60% 40% 60% 20% 80% 10% 90% Tab 2: The churn rate with the years of work Resign α.The collaborative distance between active
相关推荐：
loading