Lead Time Analysis

Improving delivery time accuracy.

Timeline

•  Sprint

Role

•  Analyst

Tools

•  Excel
•  Tableau TBD

Problem

•  Delays in production and the need to ensure on-time delivery.

Overview

A new DTC shoe brand is dealing with delays in production and we need to ensure on-time delivery by discovering what the issue is.
EXECUTIVE SUMMARY
This analysis was conducted in four parts to uncover insights with consideration to time needed on the component, product level, season, and overall levels as well as based on product line type (A, B, C, D). Based on a review of the data:
  • It is clear that there were process challenges at the factory that caused delays on more complicated tasks such as altering the standard rubber composition (D). It is suggested to consider segmenting product style D to a separate planning/demand timeline given the low quantity and the high lead times required that will likely affect the efficient and timely deployment of inventory to customers.
  • It is also suggested to conduct financial modeling to determine at what rate of comparative quantity between A-C and D keep/ return to 1 shipping drop.
  • Additionally, the average lead time remains relatively stable across several seasons, which suggests that other than the processing time required for complex products, additional delays are likely to be due, in part, to internal causes. There is a potential to dive deeper on the internal size to observe our processes and look for opportunities to streamline.
  • Finally, the small size of the shipments combined with the high delay impact creates concern. The increasing sku count could create greater risk to delivery times if waiting for the completion of the special edition footwear.
FINAL RECOMMENDATION
Consider splitting the current shipment timeline from one large batch monthly to one batch monthly of core and seasonal styles, while offloading special edition drops to once every two months to align with development time. This will reduce risk of these products affecting the delivery of the entire batch. In addition, if financially feasible, it could make production more agile.

Goal

Suggest changes to improve delivery time accuracy.
SOLVE FOR:
What is causing the delay: external or internal?
Lead Time
  • What is the average lead time? Lead time of delayed products? Per style? Per round? Max?
  • What is the average time to submit updates to the factory? Per style?
Error Rate
  • What is the overall error rate? Per style?
  • What is the correlation between error rate and long lead times?
Capacity
  • Delays per season? Internal or external?
METRICS
Lead time, error rate, inventory per season.

Lead Time: External/ Factory

Overview

This analysis was conducted in four parts to uncover insights with consideration to time needed on the component, product level, season, and overall levels as well as based on product line type (A, B, C, D). Based on a review of the data:
  • It is clear that there were process challenges at the factory that caused delays on more complicated tasks such as altering the standard rubber composition (D). It is suggested to consider segmenting product style D to a separate planning/demand timeline given the low quantity and the high lead times required that will likely affect the efficient and timely deployment of inventory to customers.
  • It is also suggested to conduct financial modeling to determine at what rate of comparative quantity between A-C and D keep/ return to 1 shipping drop.
  • Finally, the average lead time remains relatively stable across several seasons, which suggests that other than the processing time required for complex products, additional delays are likely to be due, in part, to internal causes.
TEST
  • Avg Factory Lead Time and Sku Count by Season, (Figure 1):
    to observe portfolio wide variance across those 3 factors.
  • Factory Lead Time by Product Feature, (Figure 2):
    to observe component specific issues.
  • Factory Lead Times by Product Style, (Figure 3): 
    to observe style specific issues.
  • Factory Lead Times by Product Style and Quantity, (Figure 4): 
    to surmise potential opportunity/ impact of alternate shipping options based on quantity.

Test 1: Avg Factory Lead Time and Sku Count by Season

The first part looked at average lead times and sku count by season.
Hypothesis: An increase in average lead time = an increase in the sku count at a similar rate.
Figure 1.1
Figure 1.2
ANALYSIS
  • For an efficient factory, we may expect the average lead time to remain relatively stable despite sku count changes. For a less efficient factory, we may expect a stable rate that changes to an exponentially positive rate of delay as skus increase.
  • Here we see a relatively stable average lead +/- 0.5 week, which appears to be within a healthy range of efficiency. This suggests that external issues aside from manufacturing complex components, may be due to internal factors. 
SUGGESTION
  • See next test: Analyze factors related to complex components and extreme lead times to discover opportunities for efficiency.

Test 2: Factory Lead Time by Product Feature

The second part looked at lead time based on footwear features such as printing or painting on various components, the change of standard materials.
Hypothesis: Less complex processes = shorter delays.
Figure 2
ANALYSIS
  • The total average lead time is just over 3.6 weeks, but by removing the extreme factor of products with special composition, the average lead time reduces by 0.7 weeks to 2.9 weeks.
  • Additionally, despite outlier variances driving a max lead time of 9 weeks, the typical max when removing the special composition featured products, is 6 weeks, a potential reduction of 6 weeks in the timeline.
SUGGESTION
  • The disparity is sufficiently large and the potential disparity can cause risk to deadlines--> 
    +   Can we have separate shipments for products with these features to reduce bottlenecks?
    +   Can we change the order of operations or start time to catch up to the final product assembly?
    +   Do we need to limit the quantity of products with these features to limit the risk to deadlines?

Test 3: Factory Lead Times by Product Style

The third part looked at lead times based on product style. These product styles are based on development style and pricing category with style A being the core black/white product style.
Hypothesis: higher quality/priced products = longer delays.
Figure 3
ANALYSIS
  • Products B-C have more similar average lead times, and product D has longer lead times, but also fewer products within the category.
  • Price/ quality was a factor in the lead time, but the greater impact was the use of new materials that required more time to source and test the quallity.
SUGGESTION
  • Ship products A-C monthly on a pallet at one date and ship product D every 2-3 months instead of monthly to maintain timelines for A-C and not create delays for core products.
  • Provide a longer buffer for new materials or colors and use standardization to streamline the process after approval.

Test 4: Factory Lead Times by Product Style and Quantity

The third part looked at lead times based on product style and total product quantity. Despite the clear visual of the lead times per product in test 2, it was unclear how much of an impact it would have on demand/planning if product style D was in fact offloaded to a separate shipping time.
Hypothesis: Core products (A) have the least amount of samples, while the other three styles (BCD) will have a relatively similar sample number.
Figure 4
ANALYSIS
  • Product style A accounts for 8%, but it is a core item for the catalog and product line. It also has the lowest delays. 
  • Product style B and C together account for 74% of the catalog and roughly 3.5 week delays. 
  • Product style D, a luxury item, accounts for 18% of the catalog, but has more than double the delays of styles B and C.
SUGGESTION
  • As mentioned with test 3, consider splitting the delivery times of the lower delayed products and the higher delayed products. This can reduce potential challenges with milestones, customer satisfaction, and partnerships.  

Test 4: Factory Lead Times by Product Style and Quantity

The third part looked at lead times based on product style and total product quantity. Despite the clear visual of the lead times per product in test 2, it was unclear how much of an impact it would have on demand/planning if product style D was in fact offloaded to a separate shipping time.
Hypothesis: Core products (A) have the least amount of samples, while the other three styles (BCD) will have a relatively similar sample number.
Figure 4
ANALYSIS
  • Product style A accounts for 8%, but it is a core item for the catalog and product line. It also has the lowest delays. 
  • Product style B and C together account for 74% of the catalog and roughly 3.5 week delays. 
  • Product style D, a luxury item, accounts for 18% of the catalog, but has more than double the delays of styles B and C.
SUGGESTION
  • As mentioned with test 3, consider splitting the delivery times of the lower delayed products and the higher delayed products. This can reduce potential challenges with milestones, customer satisfaction, and partnerships.