Leveraging supply chain optimization and visibility to achieve carbon reduction targets
Summary
TLDR在麻省理工学院(MIT)的供应链管理(SCM)MicroMasters课程的网络研讨会中,Laura Allegue和Kellen Betts与MIT交通与物流中心的研究科学家及讲师、可持续供应链实验室主任Dr. Josue Velazquez进行了深入对话。Dr. Velazquez分享了关于可持续供应链管理(SSCM)的洞见,特别是优化和可见性技术在推动可持续决策中的作用。他通过实际研究项目案例,展示了如何通过延长交货时间来提高运输效率并减少碳排放,同时探讨了在实施可持续供应链实践中遇到的挑战。此外,Dr. Velazquez还讨论了供应链各参与方之间数据共享的重要性,包括信任、安全和合作的考量。他鼓励MicroMasters学生利用优化和机器学习技术来解决供应链中的可持续性问题,并提供了关于如何开始这一旅程的建议。
Takeaways
- 🎓 麻省理工学院(MIT)的可持续供应链管理(SCM)项目强调了优化模型和可持续实践的重要性。
- 🌿 Dr. Josue Velazquez 讨论了可持续供应链管理中优化和可见性的作用,以及如何利用这些技术实现可持续性策略。
- 📈 通过实际研究项目展示了如何通过优化和技术创新来应对实施可持续供应链实践时面临的挑战。
- 🚚 快速配送服务的增长导致了碳足迹的增加,而通过优化配送时间和路线可以减少这种影响。
- 💡 通过提供环境影响信息给消费者,可以激励他们选择更环保的配送选项,从而减少碳排放。
- 📊 通过数据分析和模拟,研究者们能够准确估计通过延长配送时间窗口而节省的二氧化碳排放量。
- 🔍 供应链的可见性不仅涉及消费者,还包括与供应链合作伙伴共享信息,这需要建立信任和合作。
- 🤝 合作和沟通是实现供应链可持续性的关键,包括与供应商和客户共享数据和减少环境影响。
- 📝 讨论了在优化模型中包括二氧化碳排放量作为决策变量的可能性,以及如何通过算法处理这些变量带来的复杂性。
- 📈 Dr. Velazquez 建议对刚开始接触这些概念的 MicroMasters 学习者来说,应该从学习优化、技术和可见性开始他们的旅程。
- ♻️ 强调了在经济激励和环境激励之间找到平衡的重要性,以及如何通过这些激励措施影响消费者的选择。
Q & A
为什么MIT可持续供应链实验室的主任Josue Velazquez博士在这次网络研讨会上?
-Josue Velazquez博士参与了这次网络研讨会,分享了他在可持续供应链管理、优化和科技方面的专业知识。他最近在edX上推出了关于可持续供应链的大规模在线课程,并且作为MIT可持续供应链实验室的主任,他参与了多个与供应链可持续性相关的研究项目。
在网络研讨会中提到的供应链管理中,为什么环境可持续性是一个热点话题?
-环境可持续性成为热点话题是因为越来越多的公司和客户开始关注其运营对环境的影响,并设定了明确的碳减排目标。供应链管理涵盖了几乎所有商业流程,这些流程中的许多活动,如运输和材料提取,对环境有着直接的影响。
为什么在实现供应链可持续性时,优化和可见性是两个主要的工具和技术?
-优化和可见性是实现供应链可持续性的关键工具,因为它们可以帮助公司在满足客户需求的同时,最小化成本并减少环境影响。优化可以提高效率,而可见性则有助于更好地规划和决策,从而实现可持续性战略。
在网络研讨会中,Josue提到了哪些与供应链可持续性相关的实际研究项目?
-Josue提到了几个实际的研究项目,包括与墨西哥零售商Coppel合作的项目,该项目通过延长配送时间窗口来提高配送的整合性,从而减少了CO2排放。他还提到了使用优化模型来改善配送网络设计的研究。
在网络研讨会中提到的“Green Button”项目是什么,它如何帮助实现可持续性?
-“Green Button”项目是一个实验,它通过向消费者提供关于配送选项的环境影响信息,鼓励消费者选择更环保的配送方式。实验表明,超过50%的消费者愿意为了环保而延迟收货,这表明通过提高可见性和透明度,可以有效地促进消费者参与供应链的可持续性策略。
在讨论供应链可持续性时,为什么信任和合作的考虑如此重要?
-信任和合作在供应链可持续性中至关重要,因为它们是数据共享和信息透明度的基础。公司可能不愿意共享可能影响其竞争力的信息,如排放数据。因此,建立信任和明确的合作规则对于实现供应链各方的透明度和数据共享至关重要。
在网络研讨会中,Josue提到了哪些技术或方法可以用来优化供应链网络设计?
-Josue提到了多种技术,包括使用启发式算法、遗传算法、元启发式算法以及大型优化软件(如Gurobi、CPLEX、Google OR)来处理更复杂的优化问题。他还提到了多目标优化和机器学习技术在解决这些问题中的潜在应用。
为什么MicroMasters学习者应该学习优化和使用技术来帮助实现更可持续的供应链?
-MicroMasters学习者应该学习优化和使用技术,因为这些技能对于理解和改进供应链管理至关重要。这些工具和技术可以帮助他们在未来的职业中解决实际问题,提高供应链的效率和可持续性,这对于任何从事供应链工作的专业人士来说都是关键能力。
在网络研讨会中,Josue是否提到了关于供应链可持续性的职业机会?
-是的,Josue提到了供应链可持续性领域中存在多种职业机会。他提到了数据科学、物流、消费者理解以及运营研究等领域的职业,并鼓励学习者通过实践和参与相关课程来发展他们的理解。
在网络研讨会中,Josue如何描述在电子商务和快速配送环境下,供应链的挑战和机遇?
-Josue描述了电子商务和快速配送环境下,供应链面临的挑战,包括增加的碳足迹和对规划时间的压缩。同时,他也指出了机遇,比如通过提供更多的配送时间窗口,可以改善配送的整合性,减少能源消耗和CO2排放。
在网络研讨会中,Josue是否讨论了不同运输选项的价格敏感性问题?
-是的,Josue讨论了不同运输选项的价格敏感性问题。他提到了他们进行的一项研究,其中测试了消费者对于经济激励和环境激励的敏感性。研究发现,如果提供经济激励或环境激励,大多数消费者愿意等待更长时间的配送。
Outlines
🎓 欢迎与介绍
本段介绍了MITx供应链管理(SCM)MicroMasters课程的负责人Laura Allegue和Kellen Betts,他们共同主持了这次直播活动。特别嘉宾是Josue Velazquez博士,他是麻省理工学院运输与物流中心的研究科学家和讲师,也是麻省理工学院可持续供应链实验室的主任。他们讨论了关于可持续供应链管理的议题,包括优化模型、可持续实践、技术、可见性等方面,并启动了一个了解观众参与动机的投票。
📈 供应链管理与碳足迹
Josue Velazquez博士讨论了供应链管理的重要性,以及如何通过物流来减少碳足迹。他提到了公司如何通过测量和识别供应链中的“热点”来确定环境影响,并探讨了如何利用信息和决策来实现碳减排目标。他还介绍了“影响力-努力”图表,这是一种在咨询中常用的框架,用于评估不同策略的效率。
🛍️ 电商与绿色按钮项目
本段讲述了一个关于电商和快速配送服务对碳足迹的影响的研究项目。Josue博士介绍了“绿色按钮”项目,该项目通过提供环境信息给消费者,鼓励他们选择更环保的配送选项。研究表明,超过50%的消费者愿意为了环保而延迟收货。此外,还讨论了如何利用优化技术来改善配送计划,减少能源消耗和CO2排放。
🤝 供应链可见性与合作
Josue博士强调了供应链可见性在促进合作伙伴之间信任和合作中的作用。他讨论了公司如何与供应商和客户共享信息,包括排放数据和运营数据,并强调了数据共享中的挑战,如信任、安全和合作。他还提到了如何通过定义合作的范围来克服这些挑战,并实现显著的减排效果。
🌐 优化模型与可持续性
Laura询问了关于在优化模型中包含CO2排放量作为决策变量的问题。Josue博士回答说,可以将CO2排放量作为模型中的一个约束条件或目标函数,并且有多种算法和技术可以处理这种复杂性。他还提到了使用启发式和精确方法来解决这些问题,并且计算能力的提升使得解决更复杂问题成为可能。
📚 学习资源与职业发展
Josue博士为MicroMasters课程的学员提供了学习建议,推荐他们完成整个MicroMasters课程,并提到了数据科学、数据分析和机器学习等课程。他强调了在实践中应用所学知识的重要性,并鼓励学员们深入学习优化和技术分析,因为这些技能对于解决可持续性问题至关重要。
📈 网络设计与碳排放
Josue博士讨论了如何设计更接近客户的供应链网络,并考虑了由此带来的物流和环境足迹的增加。他提到了需要重新审视库存政策,并且考虑交通拥堵和车辆限制等因素。他还提到了他们的研究,这些研究帮助公司在实现商业目标的同时,也实现了碳减排目标。
💰 价格敏感性与绿色配送
Kellen提出了关于绿色按钮项目中不同配送选项价格的问题。Josue博士解释了他们如何通过调查了解消费者对不同配送选项的感知,并测试了经济激励和环境激励对消费者延迟收货意愿的影响。他指出,尽管70%的消费者表示愿意为了环境效益而等待更长时间,但在实际购买时,有52%的消费者实际上改变了他们的决定。
📝 结语与资源分享
Josue博士总结了讨论,并提供了获取他研究项目和论文的信息。他鼓励有兴趣的参与者访问sustainable.mit.edu或通过电子邮件[email protected]联系他们。他还提到了他们的社交媒体平台,并表示愿意回答更多的问题和分享他们的研究成果。
Mindmap
Keywords
💡供应链管理
💡可持续性
💡优化模型
💡技术
💡可见性
💡碳足迹
💡电子商务
💡消费者行为
💡绿色按钮项目
💡碳减排目标
💡运输活动
Highlights
MITx供应链管理MicroMasters课程负责人Laura Allegue和Kellen Betts主持了这次网络研讨会。
邀请了MIT交通与物流中心的研究科学家和讲师Dr. Josue Velazquez作为嘉宾。
Dr. Velazquez介绍了他在edX上开设的关于可持续供应链的首个大规模在线课程。
讨论了可持续供应链管理的重要性以及优化和技术在推动可持续决策中的作用。
通过实际研究项目展示了MIT可持续供应链实验室的挑战和成果。
强调了供应链管理在实现环境可持续性方面的作用,尤其是减少碳排放。
通过调查发现,大多数公司或其主要客户已经设定了明确的碳减排目标。
说明了供应链中的各个过程对环境影响的重要性,尤其是材料的运输活动。
强调了测量和识别供应链中环境活动增加的热点区域的重要性。
讨论了如何利用供应链中的信息做出更好的决策,包括努力影响图表的应用。
通过案例分析展示了延长配送时间窗口如何通过优化提高配送的整合性和效率。
介绍了优化模型在减少二氧化碳排放和燃料消耗方面的潜力和实际效果。
讨论了供应链可见性的重要性,包括与消费者和供应链合作伙伴共享信息的挑战。
强调了在供应链中实现可持续性需要技术、优化分析和消费者参与的结合。
Dr. Velazquez鼓励MicroMasters学生学习优化、技术和供应链管理知识,并将其应用于可持续性领域。
提到了在经济激励和环境激励下消费者对延迟配送的接受度。
讨论了在不同经济环境下,如何平衡价格敏感性与提供绿色配送选项的关系。
提供了MIT可持续供应链实验室的联系方式,鼓励有兴趣的参与者了解更多研究项目。
Transcripts
(bells chiming)
- Welcome everyone, and thank you for joining us today.
I'm Laura Allegue, a course lead here
for the MITx MicroMasters in SCM program.
I'm co-host in this live event with Kellen Betts,
also a course lead you may know,
and we are super fortunate to have here today
with us, Dr. Josue Velazquez.
Josue is a research scientist and lecturer
at the MIT Center for Transportation and Logistics.
He has recently launched his first massive online course
on sustainable supply chains at edX,
and today he is joining us
with one of his many hats here at the center
as the MIT Sustainable Supply Chain Lab director.
Welcome, Josue.
So before we move on, and as you know,
we love to have polls here on all our webinars.
We will kick the event with a poll.
We want to know why are you here today?
So Lisa here, who is helping us, will launch the poll,
so you'll probably see it popping up on your window.
Let us know if you're here today
because you just want to learn about optimization models
with sustainable practices,
if you just want to know more about sustainability,
if it's probably because you are here for the technology
and visibility portion of the event,
or if it is that you are part of our MicroMasters program
and you don't miss any live event.
So with that, I'll give the word to Kellen
for the agenda of the session.
- Awesome, thank you Laura, and pleasure to have you,
Dr. Josue Velazquez.
Looking forward to our conversation today.
So just kind of a quick agenda for today's webinar,
for the next 15 minutes or so,
Dr. Josue Velazquez will provide some context
on sustainable supply chain management
and the role of optimization and technology
to drive sustainable decisions.
He'll also illustrate some practical examples
from applied research projects conducted here,
the MIT CTL in the MIT Sustainable Supply Chain Lab
that he directs, and discuss the challenges faced
when implementing sustainable supply chain practices.
Laura and I will then have some questions prepared
in advance, and the last 10 minutes will definitely be saved
for your questions, and so start thinking of those,
and please use that webinar Q and A feature in Zoom,
it's that Q and A feature at the bottom there,
to ask your questions,
and be sure that you're logged in with a name.
We won't read any anonymous questions today.
We'll also share a couple more polls during our event,
so be prepared to participate with those.
And with that, let's check on the results
of that first poll here, if we can share those results.
Awesome, so the poll,
the question was why are you here today?
Lots of, it looks like lots of great learners
in our audience, so I want to learn more on how to connect
with optimization models and supply chain practices.
That's awesome.
That was 60%, and almost 40%, I'm interested in knowing more
about sustainable supply chain labs, that's cool.
I'm sure Josue will have lots of, you know,
be excited about that answer as well.
I don't know if you have any comments or thoughts
on those initial poll result there, Josue,
but with that in mind, maybe we'll just,
I'll pass the stage to you and you can kick it off,
share your slides, and take it away.
- Fantastic, well, thank you so much first, Laura,
for the invitation, and Kellen as well.
Very nice introduction,
and it's really a great pleasure to be here today.
So lemme just start the presentation with a couple of slides
that I would like to share.
All right, so all good, fantastic.
So the intention of today's discussion presentation is
to illustrate a little bit about the importance
of two main tools and techniques,
which is one, optimization, and the other visibility,
and you know, both, you know, supported by the technology,
and the idea is how we could use these methodologies,
these approaches to actually achieve
sustainability strategies, and as you know,
there are plenty of discussions around the topic.
You know, we've been working on sustainable supply chains
and sustainable logistics for many years,
and the idea is to just illustrate some of the projects
we've done in the past to also give you a glimpse
how you could use these techniques
that you potentially may also learn in this awesome courses
SE2X, SE4X, and all the MicroMasters in general
that are provided by by MIT CTL, so lemme just start.
So the first is, of course, many of you commented actually
in the poll, the importance of sustainability,
and there were three or four questions related
to this topic.
We know that it's hot topic,
we know that plaintiff of companies are starting to look
into this, and there are different dimensions,
but of course, there is particular interest
in the environmental sustainability, and we saw it.
We actually ran last year around this time a poll
with some of our corporate partners.
So what you see here on the screen is actually
for last year some of the CTL corporate partners
that also engage with us in different capacities.
And when I was invited to give a presentation
on sustainable logistics, I asked them this question.
It's either you as a company
or key customers have established
clear carbon reduction targets for the next year,
for the next 10, 20, 30, 40 years?
And then I'm giving three options,
yes, no, and not sure, right?
And I will venture to say that those that are not sure is
probably they have heard something about it,
and some probably is not still very, very explicit,
but this is what we saw, right?
The vast majority actually has either companies,
there in their strategy, or key customers
that are actually moving also,
their strategy in the supply chain,
that have a specific carbon reduction targets.
And of course, now that we know this information,
the question is why are we looking into this
from the perspective of supply chain management,
from from logistics?
And this is a key question.
To answer this, in case you are not very much familiar
with this topic, you know, supply chain management,
at least the way that we see it, actually has to do
with almost all the processes that exist
in any business in an organization.
And the way that I always like to illustrate this is to just
by looking at this picture.
So this young lady's buying a blouse or a shirt,
and there are some statistics that show
that all the processes that are required
and all the trips that this shirt has done
in the past are actually way more than what, in average,
a person like ourselves, we are gonna travel
for the rest of our lives,
so in other words, this shirt has actually traveled more,
and the reason is because of what I just explained.
There are plenty of activities related with the upstream
in the supply chain that are actually creating
a lot of transportation movements, extraction,
could be that the materials are coming from,
I don't know, cotton extracted in Texas, move in Long Beach,
send in China for assembling and dying,
maybe send again to Europe for, you know,
other type of SKU packaging allocation,
but there are different activities that are happening,
you know, so that this shirt can actually exist
in this shelf.
And those processes in general are designed in a way
that we can achieve, you know,
either the optimal ordering in terms of having
sufficient amounts to serve the demand, you know,
expectations in on-time delivery,
field rate, and service level,
but at the same time are also designing a way
that we are looking at the minimization of cost,
so we try to be as efficient as possible.
Now, all these processes again, imply different flows
of materials, which is, you know,
part of the foundation of supply chains.
But of course, the one that actually creates the impacts
for the environment are actually related
to the movement of materials.
You know, and that's why also when we look
at the supply chain and different topics on sustainability,
we tend to pay a little bit more attention
to all the transportation activities
or what are the decisions in the supply chain
that are actually driving the transportation activities.
And what I'm gonna do today is explain and illustrate
with some of these examples,
particularly for the transportation.
Now, imagine that you are this company that needs
to also serve these customers
with certain expectations and demand.
So the question is,
how can I achieve those carbon reduction targets
while at the same time,
I have to meet also my business goals?
Now, different approaches, the first one is a key.
First you need to know what is the size of the problem.
And the first step, usually companies have started
to do that probably 10 years ago,
five years more intensively, is to conduct
a carbon footprint, and this is key.
Again like in any other problem in business
and in supply chain, you get what you measure.
So looking at the different supply chain activities
and processes that are actually explaining
a specific business are key to really identify
what are the different ways that companies can operate
their supply chain, make investments in different equipment
and technology, and leverage different capabilities
to achieve their carbon reduction targets.
So first step has to do with that,
so to identify the what we call the hotspots,
so where exactly the supply chain you can observe certain,
you know, increase in environmental activity.
And by looking at this, remember this can be caused
either because you have inefficiencies in the equipment,
but could also be caused
because of all decisions you are making, for instance,
when you decide different replenishment strategies,
this immediately may affect the transportation activity
both, you know, downstream or upstream in the supply chain.
If you define different service level targets
in inventory strategies, again, you're gonna affect again
how much you're gonna keep in the inventory,
how much energy you're gonna be using in your own warehouse
or in materials handling.
So there are different processes, and it's important
in this stage to really spend time identifying
these operations.
Now, once you have this information,
then the next question is, okay,
so how I'm gonna use this information to make decisions?
And one of the frameworks that I use to teach, you know,
in the topic of sustainable supply chain management has
this famous impact effort charts that are very common
in consulting, and then you have usually here,
like those efforts that you require that is less or a lot.
If at the end, the impact is not really relevant,
don't waste your time on the those strategies.
The question is more whether you need to put a lot of effort
to achieve long-term significant reductions
to achieve your carbon targets,
or you can actually have also what we call the quick wins.
I don't need so much effort because I already have
the infrastructure, the information,
and I can actually leverage that by looking
at different decisions in the supply chain.
So as I said, one side equipment could be investment,
for instance, in electrification in the supply chain,
or in the other side looking at the different decisions.
And regardless of where is your strategy,
both of them can actually benefit extensively
from both what we call the optimization
and also the visibility, and I will give some examples.
So the right approach,
the way that we see it has to do with balancing
these two strategies.
From one side, how we can squeeze for efficiencies,
usually we try to, you know, get as much as we can
with the minimum, you know, resources we are operating,
so this is the foundation of optimization model.
And on the other side,
we look at also the consumer facing logistics,
which has to do a lot with growing the capacity
that you have to squeeze again by having visibility
and leveraging the visibility that becomes also
like a strategy to involve also customers
or consumers in your strategies.
All right, so I'm gonna illustrate with both,
and I will start actually
with the power of visibility in this example,
and hopefully by now, some of you already are aware
of the type of projects I'm presenting,
'cause I've been trying to, you know, disseminate
part of the projects we call the Green Button.
And here everything starts with understanding, for instance,
in transportation, as I said,
the importance of e-commerce and the concept
of fast shipping that started three years ago,
if I recall correctly.
The fast shipping in terms of the same day delivery promise,
before it was today, delivery promise, it was already fast,
but it became faster.
And then once Amazon launched this strategy,
as we know, all the other companies that are offering
also e-commerce is starting also both proposing a strategy
or struggling with this.
But as we know, this fast shipping and e-commerce
in turn also involves an increase in the carbon footprint.
And this is just intuition, because at the end,
you have less time to really do the planning,
which means probably you're gonna be delivering
to same neighborhoods multiple times,
either during the day or the week,
which means increasing in trips, when in fact,
you could probably wait if you will have more days
to deliver and then improve consolidation,
and then do better planning,
which in turn also reduces the amount of energy
and CO2 emissions you will have in this operation.
So for this example, we worked on a project sponsored
by by Coppel, which is one of the Mexican retailers.
And what we did is to take information
about the different last mile logistics operations,
and the idea was to look at this information
and try to first understand what are the emissions
associated with that operation.
Now, to do that, we have started working with a lot of data,
and remember, in this case, we are talking about plenty
of delivery nodes, plenty of transportation activities,
different SKUs, so what we did is to conduct
a very comprehensive estimation of the carbon emissions,
and then capture the information of what happens
every time that we delay one day the delivery.
So in other words, if I have to operate,
and I need to deliver with a window of two days,
so how the distribution center will have to operate,
and how many trucks will I need
for a specific demand that I'm estimating?
Now, if I have three days to deliver,
if I have four days to deliver,
how I will reorganize my scheduling to make sure
that I serve the customer and then I improve consolidation?
And by looking at improvement in consolidation of an amount
of trips, we were able to estimate quite accurately
what would be the amount of CO2 emissions that we will save
by having more days to deliver.
And what you're looking here at the slide is actually
a real simulation of one of the distribution centers
by looking at all this operation of the data conducted
in this last mile delivery.
Now, what we did in this case is to use this information
not just to improve, you know, squeeze for efficiencies,
but in fact, what happens if we are able to use
these estimations to actually disclose it with the consumer
at the moment of the purchase, and that's the concept.
If you know when you are buying that asking
for fast shipping will have an effect either
in the clearance of CO2 or or other type of displace
in environmental information,
will consumers be willing to delay their home delivery?
And if that happens, then we are showing in fact,
that the visibility is helping the supply chain achieve
in this case, sustainability strategies,
and this is what we did.
What I'm just illustrating here is, you know,
the short video when we were featured by CNN
on this just mock up in the e-commerce platform.
It's just a mock up, but it looks something like that.
Like you have their shipping, but then imagine
that you disclose this information and then incentivize
consumers to select the green delivery option.
And what we observed in this case that we took to cities
to do the analysis, Mexico City and Monterey,
we actually observed that more than 50% of people
in fact changed their original decision
to select now green deliveries.
This is decisions, this is not a survey.
Now, this is awesome, right?
And of course the question here is, okay,
imagine that now we actually get information
for all the different distribution centers that by providing
information, environmental information,
consumers are willing to wait five days more
for the home delivery, fantastic.
Okay, so now what do we do with this five days of delivery?
Well, this is when it comes the other side of the coin,
which is the power of optimization, and the concept is,
okay, so if we have more days to deliver,
let me just illustrate the potential of this problem.
So this, what you see in the slide is actually
two real trajectories, so in two days,
by the end of February and the first day of March,
we were having different trucks going exactly
to the same neighborhood, the same zip code.
And what you see in the nodes that are in red,
this is the deliveries for one day,
and those that are in blue are the second day.
Now imagine that you are able to deliver green, I'm sorry,
not green, blue and red any day, right?
So you can either deliver the last day, or the next day,
or actually you have even more days to deliver,
which in turn also implies you get new orders.
But let's say that you have now flexibility,
there is no same day delivery,
there is no fast shipping, and you can deliver any day.
The question is how would you rearrange this?
And then what we see here, this is just done
with a basic heuristic similar to the one that we did
in the simulation.
By rearranging, immediately we see that we are mixing
both delivering nodes, the red and the blues in both days.
We actually observe already significant reductions
in mileage, in area covered, more density,
which in turn also means much more efficiency
that we are expecting here in reduction
of fuel consumption and CO2 emissions.
So we follow this intuition and we say,
well, we realize that
traditional transportation management systems are having
plenty of overlaps because they're looking
at the minimization of distance, and what we say is like,
what if we create a different mathematical model
that instead of delivering following
the distance minimization and having overlaps,
it's actually deciding to consolidate,
wait a little bit more,
so postpone some orders so that delivery can be done later?
So we worked this project with a couple of colleagues
on the site here, and the idea was to take information
of expected orders, current orders, critical orders,
and then define a grid for delivery that will optimize
the amount of distance traveled
and the amount of stops conducted in a specific zip code.
And just to give you a glimpse, right?
For those that were saying that we are interested in knowing
more about optimization, you should be thrilled at looking
at this large scale optimization model.
So this was really a big deal, managed, you know,
in the largest settings, and we tried different ways
to model this.
We started with Gurobi,
we started later with the Google OR,
but there are different tools that allowed you to manage
a large scale setting, and in this case, as you can see,
we got more than 70 million variables overall
that were actually used to map this optimization.
The idea is by having these extended delivery windows,
how we are gonna use this information
to improve the consolidation?
And the interesting part is that we actually conducted
a pilot to compare the current proposed software
that the company was using,
versus our proposed larger scale model, and what you see
here are the absolute reductions we obtained.
From one side, what I really love is this one,
because looking at the whole cost was actually huge,
but just looking at the fuel consumption,
which is the CO2 emissions, the amount of savings is huge
compared to just squeezing efficiencies.
This is a mix of looking at visibility
and transparency plus also using optimization,
and there are plenty of examples on this.
I could spend the whole webinar just talking and talking
how you could improve your decision network,
delivery sequence, inventory sourcing, et cetera, but again,
the recommendation is you squeeze for efficiencies using
plenty of technology optimization analysis,
but on the other side, you also want to implement
some consumer facing logistics sustainability by looking
also at the power of transparency and visibility.
Anyway, appreciate your time.
I hope that you find this interesting,
and we can have now a nice discussion.
If you're more interested in knowing more
about all this content, feel free to reach out to us
in our website, sustainable.mit.edu.
Thank you so much, and back to you, Kellen and Laura.
- Awesome, well thank you, Dr. Velasquez,
a fascinating presentation of the example
of kind of showing both sides of the balance there,
if you will, you know, engaging the customer and seeing
if you can gain maybe some flexibility from the customer,
gaining more days.
That's sort of fascinating to see that 50%
of customers will choose a slower option, you know,
in today's Amazon same day delivery world,
and at the same time using an optimization
to kind of see how you can squeeze out
some of those efficiencies.
You know, fascinating discussion.
There's tons of questions that we could probably expand on,
a number of different areas, and we already have
some great questions here in the Q and A,
but maybe I want to kind of jump
into one particular question and expand on the concept
of supply chain visibility.
You know, I know you talked a lot about kind
of the customer facing side of supply chain visibility.
Some of the other side of supply chain visibility involves
engaging supply chain partners, you know,
so sharing information with, you know, upstream, you know,
suppliers, downstream customers that are other businesses,
if you will, so it often involves any receiving data
with these partners, but maybe sending and receiving,
you know, emissions data, operations data, et cetera.
But often the challenge involved with this sharing
of information, if you will, is not as much technical,
you know, it's not much how you send that data,
as much as it is the trust and the security
and the collaboration with that interaction,
with that transaction.
You know, so for example, a company may, you know,
feel that they share their emissions,
disclose their emissions data,
they'll be less competitive, for example.
So I wonder if you could discuss a little bit about how,
you know, some of these other considerations
around sharing data on this, you know,
disclosure and transparency side of supply chain visibility.
- Yeah, absolutely, no, this is very a interesting topic,
and you actually hit already in very interesting points
related to the trust and the willingness to collaborate.
So we know that plenty of the companies
at this stage, right?
Almost any any medium size or large company
in the developed world,
but also in developing countries are actually disclosing
their emissions one way or the other,
either because governmental reasons or because they want
to also be part of their carbon disclosure projects.
So you look into this, and you will find almost all
of them are disclosing their emissions,
disclosing the methodology they form.
And they commit for carbon reduction targets, but again,
for me, the question is, okay, those are absolute numbers,
then you start growing as a business,
then you report again, and it becomes an exercise
that is not that clear, particularly
if you are as a company aiming to involve customers
or consumers in your sustainability strategy.
I believe there is an important gap that at this stage,
those estimations are not translated not just into something
that can be understandable for the customer
and the consumer, but also they haven't seen
the potential to really help their own strategies, right?
Now, when they start doing it in the operational level,
this is what you were describing, it's very hard, right?
So this stage, if you are using a carrier for instance,
you are, I'm just gonna say largest cosmetics company
in the world, and then you are using
70 different logistics providers,
again the question is I need to assess performance,
but they are gonna disclose their own emissions,
and again it's gonna become a competition how they're gonna,
you know, use that information, and it can be very tricky.
On the other side, they can always say,
this is the fuel consumption I have in your operation,
but disclosing that information means they're gonna give
a strong point that can be used against them
in the negotiations, right?
If I know this is what you're spending of fuel is,
wait a minute, why you are changing me 10 times that price?
So those things become, of course, difficult,
and that's why, at least to my understanding,
it's always important to define a specific scope
of that collaboration.
So I was describing now with the consumer
and the e-commerce company, but if we look at B2B,
for instance, we've seen also examples in which a company,
you know, is looking at the carrier and it's saying,
"Can you help me?"
A shipper is saying to the carrier,
"Can you help me?
Because you need to help me achieve
my carbon reduction targets for the scope three, you know,
those emissions that are not related to the assets,
you know, that are owned by the company,"
and say, "Okay, so I can reduce the emissions,
but you guys are asking me, for instance,
to have very strict delivery windows.
So you ask me to deliver between nine to 10 all the time.
You place orders that are not consistent.
You change, you know, both frequency and amounts,
and that actually creates a more complex environment
for me to reduce emissions.
If you can help me do better planning,
if you can actually send more information,
just collaboration, those things can actually imply
significant reductions,"
probably in the magnitude of what I was just showing here
with the 30% and above.
So I believe there is potential,
and this is always a discussion,
I fully agree with you, Kellen,
that this is no more a discussion about whether we have
the technology or the software.
It's whether, you know, we both want to
and have certain rules in how to use this information,
and how we are gonna also allocate the benefits
that we are gonna be obtaining by this collaboration, right?
But this is a very interesting point
that definitely needs to be even discussed further
in the transparency and visibility topic.
- Thank you Josue, and even though we are talking
a lot about the technology in it,
I love the fact that we are talking about collaboration
and communication as a way to making our plans
more actionable and to actually be able to achieve results.
Like sometimes our goals are different
to the actions we're taking trying to get those,
so it's very important to have that in mind at all times.
So I am a huge fan of optimization,
so if you're okay with it,
I will like to go back to that part of the presentation.
In your presentation you describe how to optimize
the network design considering longer lead times accepted
by customers in order to do, as you just say,
this reduction on the carbon footprint.
And we usually use in our courses optimization for achieving
a budget restriction or a service level,
and we also know that it's possible to include a variety
of components, and we were wondering if you could share
a little bit more on whether it is possible to include
something like kilograms of CO2
as a decision variable in a model?
And with that in also in our goals,
and I would like to add on that,
that some of the members of our audience,
like Remi is also asking how can algorithms deal
with the additional complexity of adding those variables,
and if there's anything else, like I mentioned CO2,
is there any other variable that you usually include
in models when you're thinking
on optimizing towards sustainability?
- Awesome, this is great questions.
Obviously, you were my student taking my class
in sustainable supply chain management one day, but I need,
I should say the Laura actually got an A+.
But this is true, right?
So let me give the, let me start answering
some of the questions.
So from one side, your first one is,
is it possible to include an estimation
of kilograms of CO2 or environmental estimation
when doing optimization models?
The answer is absolute yes, right?
And most of the time, I'm not a fan,
even though I show that example.
I'm not that a fan of showing the kilograms of CO2
as a part of a cost objective function, a single objective,
and this is very common also in supply chain.
We do it because it simplifies, and that's okay.
We did it also because, you know,
the company was interested in looking at the cost,
but it's possible to actually measure this separately,
either as a constraint as you were describing
with the budget, say, "Okay,
so we have a carbon reduction target.
This is what we wanna have.
Now with the new design of the network, let's make sure
that all the flows we are gonna have on the location
of facilities are gonna guarantee we will not exceed
this amount of CO2,"
and then you can start playing with that.
You can, in fact, look at the model and say, "Well,
I will just change my objective function.
I will look at the minimization of CO2 emissions
instead of looking at the others,"
or you could include all of them,
you know, the multi objective optimization.
But the answer to your question is yes,
kilograms of CO2 is usually the one
that is potentially easier to include in a model
that you are aiming to include sustainability criteria,
although there are definitely many others, right?
Like for instance, last year we advised
a nice capstone project for UNICEF in Zimbabwe,
and team Kim and I co-advised that with (indistinct).
You know, these guys did an an awesome work,
like they took information to establish an objective
of minimization of kilograms of CO2 for the network design,
plus the minimization of cost,
and plus they also look at the social perspective.
They say what happens
if certain sourcing are gonna be limited
by only buying from local, you know?
So that we help the community,
and then you can change your network, right?
This is, as you probably know,
just changing certain constraints to make it work.
But you're gonna be able to measure how many
of the suppliers are actually coming from the region,
and get an estimation of the ratio,
how many of what the proportion.
And then understand also, you know,
like another social sustainability criteria.
So there're in different ways, and in general,
you just work with them with different techniques to model,
similar to what you teach in the course in SC2X for sure.
Now, you were asking me the algorithmic part.
That's the part that always, oh my,
it's always very complicated,
but there are a couple of things to say.
From one side, there is a lot of advancement on the use
of different heuristics, right?
Like, you know, genetic algorithms, meta heuristics,
there are plenty of things already exist in the literature,
and the advancement has been impressive in the last decade,
mainly also because of the advancement
of the power of computers.
Now with the new large scale softwares
like the solvers, right?
And I'm talking probably, you know better,
but there are plenty as you know, Gurobi, CPLEX,
which others?
LINGO probably, but that's, there are plenty.
I don't wanna cause here that there are ones
that I'm missing, Google OR as I was saying, right?
All of them have already some sort of heuristics coded
already, and it's possible to code them
to actually solve some of these complexities
that you are increasing because of now having more variables
and other type of objective functions.
And now the power also because, you know, in OR,
there are the guys that are fans of heuristics,
and there are the guys
that are fans of exact methods, right?
So those that are inexact methods.
They also have better techniques now,
better formulations of the model,
better ways to really conduct just typical methods coming
from the branch inbound, but actually doing even
with more power, better cuts.
So computing power has grown significantly, and I believe
now that gives us access to really solve
even much more complex problems,
and I believe we should not be scared of that.
We should, in fact, go into the high level of detail.
As you know, I'm very big fan of providing
as many data points and detail as possible
so that really the formulation of the problem captures,
to the best of the possibilities, what is exactly happening
in reality so that the solution will actually be
practical and impactful.
- Awesome, well, so many fascinating topics to jump in there
and you know, we could expand on so many of those,
and you know, even explore ideas like incorporating
machine learning and AI into some of, you know,
solving some of these optimization problems as well.
But in sake of time, Laura,
I might jump to kind of our last pre-prepared question
before we get into the Q and A.
And if you do have any questions, again, please, you know,
pop those into that Q and A feature,
and we'll we'll jump into those next.
But to just jump into our kind of last prepared question,
some of our participants here today are part
of our MicroMasters program,
and shout out to all our MicroMasters learners out there,
and they may be looking to involve themselves using
optimization or technology to help achieve
more sustainable supply chain.
You mentioned the importance of this.
Obviously, we discussed this in many different ways already
as well, but it might seem like kind of a daunting task
to a MicroMasters learner just kind of starting out
on their journey, so I'm wondering
if you could kinda maybe give some advice
to our MicroMasters learners on where they,
maybe where they should start, you know,
where they should start their journey,
learning in this program.
You know, they're just kind of learning these topics
maybe for the first time.
You know, where can they go from here?
- So those that are starting
the journey should definitely take both of your courses
and the whole MicroMasters.
That will be honestly a good start, right?
Where if you take also CRX,
you will get also an idea of what if you haven't heard
of optimization and everything else.
Now, once you have completed this,
then definitely there are plenty of ways to work on this,
and either you, you know, it is very modest at this stage,
all the data scientists that many of the companies have
and careers on that topic.
Many of other courses also on data analytics
and machine learning as you mentioned,
this is something i I probably missed,
just because it brought me back to my time
when I was doing heuristics,
but definitely, Kellen, you're totally right.
There are better ways also by combining techniques
of machine learning now to solve problems like that.
But there are careers both in practice by looking
at how to develop with certain courses,
participating in companies that are working
with this type of problems, both in logistics
as well as to understand consumers.
And on the other side, also academia, right?
So some of the students may potentially want to either join
the SCM program in some moment, or are more technical,
because SCM is more like in between, right?
Like business with engineering,
but there are others that maybe they say,
"No, I want something in operations research."
All of those things are opportunities to grow
their understanding.
So this is honestly what will be my recommendation.
There is so much information at this time available
for those that are interested in growing this understanding.
But make no mistake, getting more knowledge and abilities
in both, as I say, optimization, all the technology,
all the visibility, all the things that we are discussing,
it's something that will serve you for almost any problem.
In this case, I'm talking about sustainability,
because this is something actually quite new in a way
to be able to leverage that
to improve your strategies on sustainability.
But this could be used for almost any field, any discipline,
anybody that is working in supply chain knows
this is really, really key.
So my recommendation is embrace it, learn it, practice it,
and disseminate with others.
Tell everyone about it.
- Thank you, Josue, great piece of advice,
and definitely something that everyone
in our audience can apply no matter if they are taking
the courses or not.
I'm just willing to launch the last poll before we go
to the Q and A from the audience
so that we can let that populate.
Meanwhile, so I appreciate Lisa's help to launch that one.
This is just to know what are you taking away
from today's meeting, today's event?
What was the most interesting part for you?
That will help us a lot to get some feedback
and to see what are you into next
for our next webinar in the series.
So while we let that populate,
I want to connect one question from the audience.
It is from Arasani Med with some thought I had
while you were going through the presentation,
because Arasani is asking about other ways to ensure
sustainability in supply chains.
It's also asking about different initiatives
like probably going paperless or using renewable energy
for your manufacturing processes.
But I was thinking a little bit of e-commerce,
and you talked about a just touched upon e-commerce growth
and the challenges that may bring with the fast shipment.
And then I think with that there is a trend to be closer
to the customer, and I know you have some research on that,
and I wanted to know about the inefficiencies
and what is the cost in terms of carbon emissions
or in terms of sustainability
for an inefficient supply chain
or for a process that is not working as expected,
not only because you're not delivering on time,
but also because there is this need to return the product
back or to attend some emergencies probably.
- Yeah, no, so there are many topics that you touch,
Laura, and of course, you know, we discuss about the design
of a network that also is gonna consider returns.
You know, we should probably be talking with Eva,
Eva Ponset, right?
On those topics in closed loop supply chains
and reverse logistics, which by the way,
these are topics we definitely discuss in our edX course
on sustainable supply chain management.
As you were saying, one of the things that I definitely have
a study with the team on the lab related
to the proximity with customers, right?
With consumers, particularly in e-commerce.
When the companies, the bigger retailers launch the strategy
to do fast shipping, their main argument was, you know,
saying we are having a higher footprint,
we are getting closer to consumers,
we are gonna be able to react very fast
to unexpected changes in the demand.
And this is also what both Walmart.com, and Amazon,
and Target, many others have said.
But this study is that this cannot be just taken
as granted, because every time that you change, you know,
proximity with customers, you are definitely extending
the footprint both in the logistics side,
but also in the environmental side of your operation, right?
And this is important, like if you do it in that way,
that means you're gonna have to revisit
your inventory policy.
You're gonna probably have a little bit what we call
the waterbed effect, that by reacting very fast here,
somebody has to be prepared with you, and that also creates
more inbound transportation, unexpected changes.
So those challenges need to be considered, right?
So part of the things that we started is also to capture
the changes that we've seen in the different landscape
of delivery of cities, right?
Like when we look at the network designs, you know,
the course discuss about using the Weber model.
So I never know if I'm pronouncing it correctly,
because I believe his last name is German.
But there is a model that comes from more than 100 years
that helps understand the center of gravity approach.
And what we argue is that
the approach still definitely works,
but you need to consider that now there are streets
that even if you are assuming including distance, you know,
straight distance, now you have congestion, now you have
limitations in the type of vehicles you can access.
So again, you know, all of these things that I'm saying,
I know it sounds a little bit like,
well, how do you account for that?
Well, there are ways to really capture what has happened,
you know, in a network, and try to include that
in the models so that again, you help companies achieve
everything that they need in the business goals,
but at the same time also achieving
their carbon targets.
So I hope that, you know,
I know that your questions were plenty, Laura,
so I don't know if I was able to answer,
but at least to create a little bit of the awareness
how we deal with these issues.
- Thank you, that's fascinating.
I mean, there's so many different angles,
and I think maybe what it really does is just to show
the complexity of this space, you know?
Whether it's sustainable logistics
or sustainable supply chains more broadly,
and just the complexity of the space,
and so much opportunity to learn, to explore
from a research perspective, from a business perspective,
you know, just a fascinating space.
Awesome, so maybe, so in the interest of time,
I'm gonna, maybe we do one more question, Laura,
and then we jump into our poll results
and wrap things up here.
And so I wanted to kind of maybe expand on a question here
from Linda Lynn, and expanding on it, you know,
in the context of we're obviously
in a very challenging pricing environment.
Obviously, have the inflation running high,
you know, everywhere.
You know, we're now we're facing maybe like, you know,
economic slowdown in some places
or economic slowdown globally.
At the same time, you know, when you present,
you know, different options to customers,
price is kind of a key feature
of that decision making process, right?
So her question is about the Green Button, you know,
project that you explored and whether or not
the two different shipping options that you presented were
the same price, and what is, I know there's kind of a number
of different angles we could take on that concept
of what are some of the different price sensitivities
versus shipping speeds versus, you know,
carbon footprint shipping options
that the research is showing in that kind of space.
- Yeah, absolutely, absolutely, Kellen.
Thank you for the question.
So in the example that I show in the slides is
merely an illustration of how it looks the mock up, right?
But we actually did a much more comprehensive analysis.
So from one side we started the project surveying
first 1,000 approximately consumers, right?
During the home delivery operation,
and part of the questions was first to ask
in a specific region what was their perception
of the delivery, whether it was fast as low or normal?
Without really saying more we'll say, well, what if now
I need to deliver maybe with more days,
like three or four days,
would you have been willing to wait longer?
And there are some that actually responded yes, no,
because some actually were saying you deliver on a Wednesday
and then you actually, I was not even at home,
and I opened the package from the weekend,
which you know, we usually all do that, right?
We push for fast shipping, we have it at home,
and we take five days to actually open it, if not more.
So we detected that, and then later we actually asked
about economic incentive.
They said, "Well, what if I give you some, you know,
gift card or certain money
for you to wait more days for the delivery?"
And then finally we asked,
well, what if I tell you that every day you delay
the delivery, I'm able to to have a specific savings
in environmental impact?
And we actually tested different ones.
So the project is, you know, we test CO2,
we tested trees, we tested all other things.
Now, what we saw is that first people are still are
very sensitive to economic incentives, right?
And you were talking about the pricing and everything.
For sure 71% of consumers claimed
they would have been willing to wait longer
if given an economic incentive.
Now, but in that survey, we also asked
about environmental incentive, and what happened is that
we didn't get 71, but we got 70%.
So 70% of customers claimed I would have been willing
to wait longer if you will have given me the information
of the environmental incentive.
Now, what we know is that while the 70 is not the 52
that I show, but the 52 was at the moment of the purchase,
and we didn't have any, both were same,
like it was for free to deliver, right?
Like they already were members or whatever is the scheme
that they're using, but it was not one more expensive
or cheaper, right?
Because by the way, if the case of green deliveries if
still for me suspicious,
why somebody will actually charge more, right?
Particularly if they are delaying, right?
If you delay, you actually as a company,
you are saving a lot of money as I just show in the model.
The question is how you also share that.
Maybe you are saying I'm willing to put part of that money
to you know, offset certain carbon so it's, you know,
finance some projects, who knows?
But yeah, my point is we have studied plenty, right?
And we detected first that customers definitely are
sensitive to economic incentive.
We also detected they claimed to be
to the environmental incentive.
We actually captured 52%.
They actually changed their decision to delay,
which is awesome, but of course again,
that experiment was done before all these issues, right?
And I argue that, we argue that those results for sure are
contingent to plenty of things.
If you're buying things because of urgency,
could be because of hopefully there is a wedding
or a party you need to buy a present, could be maybe,
you know, in a pharmaceutical context,
something that you really need, right?
So you're not gonna be willing to wait.
There are others that maybe you're saying I'm buying coffee,
I don't mind to wait a little bit longer.
So depending on the type of product,
the purpose of the purchase, you know,
a lot also of your profile as a consumer,
so whether you have time or not to check
all the delivery options when you are buying,
so there are plenty of different conditions
that definitely need to be discussed.
And in this project we have some settings,
and those that are interested,
hopefully will see a working paper very soon available
to analyze at least our context.
But all those questions you are commenting, Kellen,
are definitely relevant and it definitely,
it matters the context, the audience, the conditions
in which you are running this analysis.
- Thank you Josue, and in the interest of time,
I would like to share the results of the poll,
and just let you know that there is plenty of questions
unanswered, and we are going to share those with you
afterwards just so that you see
what's the audience interested in.
And the interesting part of the answer to the poll is that
most people here got what we promised.
So they are expanding knowledge on sustainable supply chain,
and they found that interesting.
They also learned about the use of technology and visibility
to improve supply chains, and that's awesome.
So thank you for bringing those insightful presentation
and discussion comments to us.
I would like to give you the word,
and let you say some final words.
They are very interested in knowing where to find
all your research projects or your papers.
So if you want to repeat again your website,
that would be awesome.
- Fantastic, yes, no, absolutely.
So if you're interested, sustainable.mit.edu,
and we should probably have also
so both will take you to the same one, and yeah,
we have also plenty of social media
if you can try and find us everywhere.
We have, you know, LinkedIn, Instagram, everywhere.
Feel free to reach out and connect,
and we would be more than happy to expand
on your questions as well as the research we've been doing.
But no, I mean, just to wrap up from my side,
Laura, Kellen, really a great pleasure to be here with you.
I enjoyed very much your questions, and also sharing part
of the things that we are doing, and I look forward
to keep engaging and working together
in this MicroMasters course as well as the new edX course,
and thank you again for the opportunity.
Wish you best of luck.
- Thank you, thank you Josue,
thank you, Kellen, for a great discussion.
And everyone in the audience,
the winter webinar series is just starting,
so we hope to see you all soon in the next live event
that Kellen and I will be co-hosting, right, Kellen?
- Awesome, looking forward to it.
Thank you Laura, and thank you Dr. Velaszquez.
- Thank you. - [Laura] Thank you, everyone.
Stay tuned.
Bye-bye. - Bye.
(bright music)
Weitere verwandte Videos ansehen
From Automated to Autonomous Supply Chains
The Critical Role of Supply Chains in Business and Society
How Can Supply Chains Embrace Sustainability?
Sustainability in Pharmaceutical Industry - Moderna
Lean, Green, and Digital: Leveraging SCM Tools to Balance Complex Tradeoffs
Creating customer value using SCM retail technologies and systems
5.0 / 5 (0 votes)