ISyE 6203 Project Dynamic Allocation of Labor in

ISyE 6203 Project Dynamic Allocation of Labor in

ISyE 6203 Project Dynamic Allocation of Labor in a Distribution Center : Staples Team Aravind Meenatchi Sundaram Anirudh Ramkumar Ajay Balan Saravanan Pitchaimani Siddhesh Wavhal Deepak Gahlawat ISYE 6203 Project Staples is a global company with delivery and retail businesses serving customers in 27 countries With $25 billion in sales, it serves businesses of all sizes and consumers in 26 countries throughout North and South America, Europe, Asia and Australia. The Three business segments are: North American Delivery, North American Retail and International operations. Staples operates 1,872 retail stores throughout the United States and Canada, offering savings on more than 7,000 office products. Project Background Staples.com is an award winning e-commerce site that features 30,000 products ranging from basic office supplies to the latest business technology. With 29 fulfillment centers and a fleet of about 820

trucks, Staples is able to offer fast, friendly delivery throughout North America. Their DCs serves these fulfillment centers: Employs Associates as Full time and Contract basis. These associates work across 12 Departments and some of the associates are cross trained. Project Background Departments Wave Replenishment Lean Replenishment Bulk- Full Case Reserve Bulk Pick to Belt Bulk Non Conveyables Bulk Bulk Wave Bulk Furniture Picking Break pack/ KIVA Picking QC/Finalizing Shipping Wrap & Label Returns Receiving

Put away Project Background Forecasted Demand 30000 25000 20000 15000 10000 5000 0 l l ng ay nt nt ve elt les ve ng ng ng ng e ns ro ie vi -aw me hme eser to B yab Wa icki icki alizi ippi Lab etur ont c t sh s e k P P & R C Re Pu eni leni se R Pick nv Bul ure IVA /Fin Sh ap y l o r a C or t t p ep l C k - C k ni /K Q e W n n l r k l l ve e R n R u u o Bu Fu pac In av Lea k - F B k - N k l lk ea l

W Bu Bu Br Bu Productivity (bph/hr) 700 600 500 400 300 200 100 0 l l ng ay nt nt ve elt les ve ng ng ng ng e ns ro ivi -aw hme hme ser to B yab Wa icki icki alizi ippi Lab etur ont e e c t s s P P & R C e k Re Pu leni leni se R Pick nv Bul ure VA /Fin Sh ap ry I o r a C t p ep l C k - C k - ni /K Q to e W n l n k l l r

ve e R n R u u o Bu Fu pac In - ak av Lea k - F B k - N k l l e l W Bu Bu Br Bu Project Background Current Associate Information Full time $17.50/hr 8 hours per day Part Time $12.50/hr 6 hours per day Overtime $24 / hr Limited to 4 hr/person Only Full time Associated can perform overtime Objective Achieve optimal staffing levels and labor allocation in their Distribution Centre. Maintain flexibility to serve variable demand that the DC faces over a year. Minimize Labor Cost Methodology PART 1 ~ Over an year Optimal Staffing Level for each department over a

year 1. Balance number of associates on Full time and part time maintained on payroll to minimize staff levels PART 2 ~ Over a day Using the staffing level obtained in part1, 1. Determine the optimal labor allocation in a day 2. Divide the days work among various labor categories. Labor categories 1. Full time Specialized 2. Full time cross trained 3. Full Time Over time 4. Part time Associates PART 3 ~ During a day Using the optimal labor achieved over a day 1. Determine number of employees to be called during each period in a day. a. Period spans for 30 minutes b. Working hours vary by department Deciding the long term labor levels Methodology: Using a Newsvendor model to 1. Calculate the optimum number of labor hours to be employed per day, so as to 2. Balance the over-utilization and underutilization of labor, while 3. Considering the variability of the demand for a

period of one year so as to minimize the annual labor costs Methodology Split the total number of labor hours to be employed between full time and part time according to a practical ratio between the them 1. This ratio is based upon the no show probability of the part time workers based upon historical data Problem Formulation For each department CU = Shortage cost = rate($/hr) in overtime rate($/hr) in regular time CO= Overage cost = rate ($/hr) in regular time Critical ratio = CU /(CU + CO) We calculate the required labor hours, based on this critical ratio and the CDF of the demand data Cost Calculation for each day If the number of hours is lesser then the critical value, then Cost = Critical Value * Regular time rate Else if the number of hours is greater than the critical value Cost= ( Critical Value * Regular time rate) + ((Hours usedCritical value)*overtime rate) Long Term Labor Allocation Tool Functionality 1. Determines the long term labor levels for each departments 2. Comparison of the existing labor plan with the optimized labor plan, considering similar parameters 3. Provision for determination of future staffing levels according to a pre-specified growth % Inputs

Daily Demand Data for each department for a year Labor wage details Working hours per day No. of workers employed in the current plan Ratio of full time to total laborers Anticipated annual growth % Output Optimized Labor levels Previous Labor Allocation Labor hours that shoud be employed per day in the long run- NEW SCHEME Departments Receiving Wave Replenishment Lean Replenishment Breakpack Bulk total QC Shipping Inventory Control Labor Hours used - CURRENT SCHEME Regular time labor hours Full time labor hours Part time labor hours 180 68 142 420 106 88 145

70 169 68 142 303 94 68 131 70 Departments 11 0 0 117 12 20 13 0 Regular time labor hours Full time labor hours Part time labor hours Receiving Wave Replenishment Lean Replenishment Breakpack Bulk total QC Shipping Inventory Control 212 80 128 388 108 82 132 80 200

80 128 280 96 64 120 80 12 0 0 108 12 18 12 0 Annual cost of Labor: New Scheme Departments Receiving Wave Replenishment Lean Replenishment Breakpack Bulk total QC Shipping Inventory Control Total Annual Cost $793,170 $356,677 $639,792 $1,756,706 $474,770 $387,488 $620,349 $338,553 $5,367,506 Total Annual cost under our scheme Total annual cost under earlier scheme

$5,367,506 $5,470,127 $5,480,000 $5,460,000 $5,440,000 Annual Cost of Labor: Old Scheme Departments Receiving Wave Replenishment Lean Replenishment Breakpack Bulk total QC Shipping Inventory Control Total Annual Cost $5,420,000 $5,400,000 $842,859 $363,665 $650,602 $1,775,200 $475,699 $390,989 $626,563 $344,549 $5,470,127 $5,380,000 Series1 $5,360,000 $5,340,000 $5,320,000 $5,300,000 Total Annual cost under our scheme

Department-wise Cost Calculations Total annual cost under earlier scheme Results The overall savings that we achieved with this model is $0.1 million or 2% of total labor costs The optimized labor levels vary only slightly from the current labor levels This shows that the current long term labor allocation is pretty accurate and no major changes are recommended Usability This tool is recommended to be used yearly or quarterly, according to the managers requirements The data required for this tool is readily available, hence there is an ease of use Input data would be fed from DOR(Daily Operating Report) in the same format Limitations This tool is conservative. It does not account for changes in labor allocations done daily or weekly as per the demand forecasts which is considered in the subsequent model 300 250 200 Receiving and put away Regular time assigned Hours assigned to Full time labor 150 100 50 0

1 9 17 25 33 4 1 49 57 6 5 73 8 1 8 9 97 0 5 1 3 2 1 2 9 3 7 4 5 5 3 6 1 6 9 7 7 8 5 9 3 0 1 0 9 1 7 2 5 3 3 4 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 Part 2 Objective: Using the staffing level obtained in part1, 1.Determine the optimal labor allocation in a day 2.Divide the days work among various labor categories. Methodology: 3.Based on the demand forecasted for the day 4.Linear programming assignment model was built in excel to determine the optimal labor allocation Assumptions: 5.Productivity was assumed to be the same for all the associates 6.Pay rate was assumed to be fixed for different labor categories. Inputs Department Receiving Put-away Wave Replenishment Lean Replenishment Bulk - Full Case Reserve Bulk - Pick to Belt Bulk - Non Conveyables Forecasted Demand FT Count PT Count 25000 13 10 25000 9 5 3000 9 5 11072 14 5 700 2

5 3000 2 5 5873 2 5 Bulk - Bulk Wave 7477 2 5 Bulk - Furniture Picking 600 10000 10000 27207 1845 2000 8000 2 31 7 8 1 5 9 5 23 5 5 5 5 5 Breakpack/KIVA Picking

QC/Finalizing Shipping Wrap & Label Returns Inventory Control Across BP/QC CT Count 4 3 4 10 Across Ship/Wrap/Returns Inventory 4 Across all departments 4 Cross Trained Receiving/Put Away CT - 1 CT - 2 Across Replenishments Across Bulk operations CT -1: Cross Trained between specific departments CT-2: Cross Trained Across all Departments Part 2: Model Constraints: Objective: Min Allocated Labor hours= Number of labors assigned * cost

Assignment Constraint: For all department, No. of Full time Specialized Employee assigned = No. of Full time employees available Number of Cross trained Employees Assigned = Number of Cross Trained employees available No. of part time Specialized Employee assigned No. of Part time employees available Demand Constraint: For any department i, Number of Associates * Hours Employed * Productivity Demand Forecasted All Full time associates would be assigned . Trade off is established between the part time associates Part 2: Results Achieved Optimal allocation of labors in a day based on the demand forecasted. Department Receiving Put-away Wave Replenishment Lean Replenishment Bulk - Full Case Reserve Bulk - Pick to Belt Bulk - Non Conveyables Bulk - Bulk Wave Bulk - Furniture Picking Breakpack/KIVA Picking QC/Finalizing Shipping Wrap & Label Returns Inventory Control Department Receiving Put-away FT 13 9 9 14 2

2 2 2 2 31 7 8 1 5 9 116 Total Labor Count PT CT-1 0 1 0 3 0 0 4 3 0 0 1 0 2 4 0 0 1 0 20 4 5 6 5 0 0 0 5 4

4 0 47 25 192 CT-2 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 4 OT mins 0 0 0 27 0 168 71 0 0 0 0 275 187 531 0 1259 Part 3

Objective: Using the optimal labor achieved over a day Determine the number of employees to be called during each period in a day. a. Period spans for 30 minutes b. Working hours vary by department Methodology: 1. Demand that occurs over the day is stochastic in nature , forecast is 95% accurate. 2. Each of the department is allocated various categories of employees based on its demand. 3. A full time and part time associate would work for 8 hours and 6 hrs continuously.. Part 3: Model Constraints: Objective: To assign associates across different time periods Assignment Constraint: For all department, No. of Full time Specialized Employee assigned = No. of Full time employees allocated from part 1 Number of Cross trained Employees Assigned = Number of Cross Trained employees allocated No. of part time Specialized Employee assigned = No. of Part time employees allocated Demand Constraint: For any period i, Number of Associates * Hours Employed * Productivity Demand Forecasted period All Full time associates would be assigned . Part 3: Result Demand Full Time Specialized Full Time Cross Part Time 25000 Assigned Workers Assigned Hours 13 8 1 8

0 6 OT hours 0 Time % Work Done 8:00 8:30 9:00 9:30 Output 10:00 10:30 11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00 15:30 16:00 16:30 Legend Blue: Input Green: Output Input to the Model 17:00 17:30 18:00 18:30 19:00 19:30

20:00 20:30 21:00 4 5 5 5 3 3 3 3 3 3 3 4 4 5 6 7 7 8 8 6 5 0 0 0 0 0 0 FT SpecializedFT Cross TrainedPT Specialized Extra Hours Total Workers Total Work Accomplished 0 0 0 0 0 0 0 0 0 0

0 0 2 0 0 0 2 230 1 0 0 0 3 345 1 0 0 0 4 460 2 0 0 0 6 690 0 0 0 0 6 690 0 0 0 0 6 690 0 0 0 0 6

690 1 0 0 0 7 805 1 0 0 0 8 920 1 0 0 0 9 1035 1 0 0 0 10 1150 1 0 0 0 11 1265 0 1 0 0 12 1380 0 0 0 0 12 1380

2 0 0 0 14 1610 0 0 0 0 14 1610 0 0 0 0 12 1380 0 0 0 0 11 1265 0 0 0 0 10 1150 0 0 0 0 8 920 0 0 0 0 8 920 0

0 0 0 8 920 0 0 0 0 8 920 0 0 0 0 7 805 0 0 0 0 6 690 Part 3: Result Receiving Receiving Demand Putaway Total Work Accomplished Putaway demand 2500 1600 1400 1200 1000 800 600 400

200 0 8:00 2000 1500 1000 500 0 8:00 9:12 10:24 11:36 12:48 14:00 15:12 16:24 17:36 10:24 BreakPack BPH Demand Total Work QC Demand 600 500 500 400 400 300 300 200 200 100 100 9:36

12:00 14:24 16:48 15:12 17:36 20:00 22:24 Quality Control 600 0 7:12 12:48 Total Work Accomplished 19:12 21:36 0:00 0 8:00 10:24 Total Work Accomplished 12:48 15:12

17:36 20:00 22:24 Summary/Conclusion 1. Optimal number of Full time and Part time associates to be employed over an year was determined. 2. Excel models were built to determine the optimal number of employees to be called-in during a day based on the demand forecasted. 3. Also the assignment of the employees across various time periods during a day was arrived. 4. This model assures better distribution of cross trained employees across department. 5. This model can be used to dynamically allocate labors across days based on the demand forecasted Future Steps/Recommendation 1. Including the benefits provided for full time employee in their pay rate 2. Accounting the reliability of the part time workers in terms of their productivity. 3.

Accounting for the cost involved in cross training. 4. Explicitly displaying the benefits to each department that would encourage in cross trained worker sharing. 5. Seasonal demands at the end of the year can be met by hiring additional part time associates. 6. Determining the optimal number of cross trained employees would minimize cost while maintaining the required flexibility to meet demand. u o Y k n a h T ? ? ? s n o ti s e u Q

Recently Viewed Presentations

  • Chapter 15 Digestion and Nutrition 1 CopyrightThe McGraw-Hill

    Chapter 15 Digestion and Nutrition 1 CopyrightThe McGraw-Hill

    Mouth. A. The mouth is the first portion of the alimentary canal; it functions to receive food and begins mechanical digestion by mastication. 1. The . oral cavity . is the chamber between the palate and tongue. 2. the narrow...
  • The American Dream - Weebly

    The American Dream - Weebly

    The American Dream . Many people immigrated to America in search of a new life and to escape persecution or poverty in their homeland. Immigrants dreamed of making a fortune...but for many, this dream became a nightmare…
  • Winnetka - Wheat Ridge, Colorado

    Winnetka - Wheat Ridge, Colorado

    To try to encourage response, we have multiple contacts with each household: a prenotification postcard, followed by a survey packet with a postage-paid reply envelope, and a second survey mailing with a reminder letter and another copy of the survey.
  • Goals of IEEE Presentation Visualizing Sustainable Solutions

    Goals of IEEE Presentation Visualizing Sustainable Solutions

    Is a faster 5G phone or block chain app going to save us? ... 1/3 of reef corals. This is the WEB that enables all life if survive together. ... and education aren't getting the job done. We need an...
  • all caps title 48 pt - Mapam

    all caps title 48 pt - Mapam

    A new, easy way to view sample Cigna ID cards. View sample ID cards for certain plan types. Choose the image that matches the patient's ID card. See both the front and the back of the card. Choose "About This...
  • September 4, 2009 HITSP 2009 TC Timeline Overview

    September 4, 2009 HITSP 2009 TC Timeline Overview

    September 4, 2009 HITSP 2009 TC Timeline Overview Note: Quality Measures, Consumer Preferences, Clinical Research, Common Data Transport may deviate from the timeline below, as their timelines are further established they will be included as separate line items
  • Batch Conflict Resolution - unipd.it

    Batch Conflict Resolution - unipd.it

    Outline The challenge of massive M2M access Random access with MPR and SIC Approximate throughput model Asymptotic analysis Conclusions ITA - 2015 * Challenges for M2M access Massive number of users Sporadic traffic Short messages Current access schemes are not...
  • Section 6.3 Periodic Trends - Mesa Public Schools

    Section 6.3 Periodic Trends - Mesa Public Schools

    } Radius #1. Atomic Size - Group trends As we increase the atomic number (or go down a group). each atom has another energy level, so the atoms get bigger. H Li Na K Rb #1. Atomic Size - Period...