HW 5 – Business analytics

Java Python HW 5 – Business analytics

This homework is due before class 6. Please submit two files: your write-up and your Excel file. If you make any additional assumptions, state them clearly.

To solve the problem, you will need to do the following:

A. Use the Generalized Analytics Procedure (GAP) to set up your problem as follows:

i. Define your model in words

1. Identify the firm’s/manager’s objective function in words

2. Identify the decision variables in words

3. Identify the random variables (risk sources)

4. Identify the constraints (optional here)

ii. Formulate your model mathematically

1. Define the decision variables

2. Define the random variables (risk sources). What is the probability distribution of those random variables?

3. Define objective function in terms of decision variables and random variables.

4. Define the constraints (optional here)

iii. Solve the problem in Excel

1. Generate MANY (>1000) random draws from the specified distribution (see step ii.2 above)

2. For each random draw calculate the objective function value

3. Try different values for your decision variable and choose the value of decision variable that results in the highest objective function value, on average.

B. Answer the questions stated in the problem (in words).

Note: I recommend starting with the GAP (Steps i and ii above). However, if you prefer to skip the GAP and go straight to Excel, points will not be deducted.

Beyond Armor

The Baltimore based com HW 5 – Business analytics pany Beyond Armor (BA) is exploring a new business opportunity: selling custom screen-printed sweatshirts for college football bowl games. BA is trying to determine how many sweatshirts to produce for the upcoming Tangerine Bowl game. During the month before the game, BA plans to sell their sweatshirts for $30 each. At this price, they believe the demand for sweatshirts will be uniformly distributed between 5,000 and 15,000.

One month after the game, BA plans to sell any remaining sweatshirts to the local TJ Maxx and Marshalls outlets for $12 each. At this price, BA believes they will be able to sell either 500 units with probability 30%, or 750 units with probability 40% or 1000 units with probability 30%.

Any remaining sweatshirts will be donated to a local charity.

BA can order custom screen-printed sweatshirts for $10 per sweatshirt in lot sizes of 200. Use simulation modeling to answer the following questions.

(a) Determine the expected profit that BA would earn if they ordered 10,000 sweatshirts.

(b) How many sweatshirts would you recommend BA order to maximize expected profit? Use the “data table” function in Excel to find the optimal order quantity.

(c) Due to an outbreak of a novel infectious disease, the governor has announced that there is a 50% chance that all sport events will now be held without a live audience. If that happens, BA will not be able to sell any sweatshirts for $30, and instead will only be able to sell to TJ Maxx and Marshalls, (in the same quantity as in the original problem formulation). How many sweatshirts would you recommend BA order to maximize expected profit? Is the venture still profitable?

Note that BA makes their order quantity decisions before they find out whether sporting events are allowed to be held.

Use the “data table” function to find the optimal order quantity.

(d) Use your calculations in part c) to create a plot of the expected profit as a function of order quantity. The plot should show order quantity on X-axis and expected profit on Y-axis         

【负荷预测】基于VMD-CNN-LSTM的负荷预测研究(Python代码实现)内容概要:本文介绍了基于变分模态分解(VMD)、卷积神经网络(CNN)和长短期记忆网络(LSTM)相结合的VMD-CNN-LSTM模型在负荷预测中的研究与应用,采用Python代码实现。该方法首先利用VMD对原始负荷数据进行分解,降低序列复杂性并提取不同频率的模态分量;随后通过CNN提取各模态的局部特征;最后由LSTM捕捉时间序列的长期依赖关系,实现高精度的负荷预测。该模型有效提升了预测精度,尤其适用于非平稳、非线性的电力负荷数据,具有较强的鲁棒性和泛化能力。; 适合人群:具备一定Python编程基础和深度学习背景,从事电力系统、能源管理或时间序列预测相关研究的科研人员及工程技术人员,尤其适合研究生、高校教师及电力行业从业者。; 使用场景及目标:①应用于日前、日内及实时负荷预测场景,支持智慧电网调度与能源优化管理;②为研究复合型深度学习模型在非线性时间序列预测中的设计与实现提供参考;③可用于学术复现、课题研究或实际项目开发中提升预测性能。; 阅读建议:建议读者结合提供的Python代码,深入理解VMD信号分解机制、CNN特征提取原理及LSTM时序建模过程,通过实验调试参数(如VMD的分解层数K、惩罚因子α等)优化模型性能,并可进一步拓展至风电、光伏等其他能源预测领域。
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
红包 添加红包
表情包 插入表情
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值